Highlights
This large validation study suggests that current guideline-based fibrosis screening criteria are too broad and capture a very large fraction of adults, many of whom have low likelihood of clinically significant liver fibrosis.
A refined, data-driven strategy increased the prevalence of elevated liver stiffness among those selected for screening while reducing the screened population to roughly one-tenth to one-fifth of adults.
In long-term UK Biobank follow-up, individuals who met the refined criteria had substantially higher risks of cirrhosis and liver-related death, supporting the clinical relevance of the selection approach.
Study Background
Chronic liver disease is an increasingly important public health problem, driven in large part by metabolic dysfunction-associated steatotic liver disease, alcohol-related liver disease, and viral hepatitis. Liver fibrosis is the key prognostic feature: once fibrosis becomes advanced, the risks of cirrhosis, portal hypertension, hepatocellular carcinoma, liver transplantation, and liver-related death rise sharply. Because fibrosis is often clinically silent, many guidelines recommend case-finding or screening strategies based on common metabolic risk factors such as obesity, diabetes, and abnormal liver enzymes.
The challenge is that broad screening criteria can generate an enormous eligible population. This creates practical barriers in primary care and specialty pathways, including high testing volumes, downstream costs, and uncertain yield. The study by van Kleef and colleagues addresses this implementation problem directly: can fibrosis screening be targeted more efficiently without missing too many patients at meaningful risk?
Study Design
This was a population-based derivation and validation study with linkage to clinical outcomes. The investigators first analyzed adults aged 18 to 80 years from the 2017-2020 National Health and Nutrition Examination Survey (NHANES), restricted to individuals with body mass index of at least 18.5 kg/m2 and without excessive alcohol use. The derivation cohort included 5,904 participants. Liver stiffness measurement (LSM) by elastography served as the key noninvasive marker, with LSM ≥8 kPa used as a threshold for increased fibrosis risk.
The authors examined how many people would qualify for screening under current guideline-defined indicators and what the prevalence of elevated LSM would be among those eligible versus not eligible. They then derived refined screening criteria that targeted only subgroups with at least a 10% prevalence of LSM ≥8 kPa. These refined criteria were tested in an independent pooled validation cohort of 13,295 adults and further linked to long-term liver outcomes in the UK Biobank, which included 282,852 participants.
The main outcomes were prevalence of elevated LSM in screened versus unscreened groups and, in UK Biobank, 10-year risks of incident cirrhosis and liver-related death.
Key Findings
Current guideline-based screening captures most adults
In the NHANES derivation cohort, 8.9% of participants had LSM ≥8 kPa. However, the screening criteria embedded in current recommendations identified 60% to 76% of the general adult population as eligible for fibrosis screening. Among those who met these broader criteria, the prevalence of LSM ≥8 kPa was only 11% to 14%. Among those who did not meet criteria, the prevalence was still low at about 2%.
These findings show the central inefficiency of the existing approach: many adults are flagged for screening, but only a modest fraction actually have elevated stiffness suggestive of significant fibrosis risk.
A refined strategy substantially reduces the screening pool
Using a data-driven threshold of at least 10% prevalence of LSM ≥8 kPa within a subgroup, the investigators developed a more selective strategy. This refined approach reduced the proportion of adults considered eligible for screening to 22% in the derivation cohort. Importantly, the prevalence of LSM ≥8 kPa among those selected for screening rose to 28%, while remaining low at 4% among those not selected.
In the pooled validation cohort, only 14% met the refined criteria, showing that the approach retained strong selectivity across independent populations. The refined strategy therefore appears to be much more efficient than existing broad recommendations, at least from a population-targeting perspective.
Performance was consistent across fibrosis thresholds
The refinement remained directionally consistent when alternative LSM thresholds were examined. In the validation cohort, among screening-eligible individuals, the prevalence of LSM ≥8, ≥10, and ≥12 kPa was 22%, 13%, and 8%, respectively. Among those not eligible, the corresponding prevalence values were 4%, 2%, and 1%.
This gradient is clinically important. It suggests that the refined criteria not only enrich for mild-to-moderate stiffness elevation, but also track with progressively higher thresholds that are more suggestive of advanced fibrosis.
Screening eligibility also predicted liver-related outcomes
The most clinically persuasive validation came from the UK Biobank outcomes analysis. Individuals meeting the refined criteria had substantially higher 10-year risks of incident cirrhosis and liver-related death than those who did not meet criteria. Specifically, the 10-year risk of incident cirrhosis was 0.99% versus 0.18%, and the risk of liver-related death was 0.40% versus 0.07%.
These differences represent roughly a six-fold increase in risk. While the absolute risks remain relatively low in a general population cohort, the separation in outcomes supports the idea that the refined criteria do not simply enrich for abnormal imaging findings; they identify a group with more meaningful downstream clinical risk.
How the results should be interpreted
The study does not argue that fibrosis screening should be abandoned. Rather, it provides evidence that the current net is likely too wide. A more selective approach could improve the positive yield of screening, reduce unnecessary testing, and better align limited healthcare resources with patients most likely to benefit from further assessment.
That said, a narrower strategy also raises the classic trade-off in screening science: fewer people tested means less burden and better efficiency, but potentially a risk of missing some individuals with fibrosis who fall outside the selected criteria. The reported 4% prevalence of LSM ≥8 kPa in the non-eligible group suggests that some at-risk patients would still be missed, though the clinical importance of those missed cases depends on progression risk, competing mortality, and feasibility of repeated reassessment over time.
Expert Commentary
This work speaks directly to a central implementation gap in liver disease prevention: we have more candidates for fibrosis screening than most systems can realistically evaluate. Guidelines have generally prioritized sensitivity, often by recommending screening in adults with obesity, diabetes, metabolic risk factors, or abnormal liver tests. The present study shows that this strategy may be overly inclusive in real-world populations.
The major strength of the study is its multiphase design. The authors used a nationally representative US cohort for derivation, an independent pooled cohort for validation, and a very large biobank dataset to connect screening eligibility with hard clinical outcomes. That combination strengthens both internal consistency and translational relevance.
There are also important limitations. Liver stiffness by transient elastography is a validated noninvasive marker, but it is still a surrogate rather than histology. LSM thresholds of 8, 10, and 12 kPa are useful for risk stratification, yet they are not interchangeable across etiologies, devices, or clinical contexts. NHANES and UK Biobank populations may not fully represent more diverse healthcare settings, particularly regions with higher prevalence of viral hepatitis, heavier alcohol exposure, or different sociodemographic risk profiles. In addition, screening performance must be judged not only by prevalence enrichment but also by whether earlier detection changes outcomes through effective treatment pathways.
Another issue is implementation. A refined screening strategy will only improve care if primary care and specialty systems can apply it consistently, access second-line noninvasive testing, and ensure timely referral for patients with high-risk results. Otherwise, better targeting may simply shift inefficiency downstream.
Despite these caveats, the study is an important contribution to the growing effort to make liver fibrosis screening more efficient and more scalable. It reinforces the notion that screening should be risk-stratified, not universal within broad metabolic categories.
Clinical Implications
For clinicians, the key message is pragmatic: broad guideline-defined eligibility may overestimate who truly needs fibrosis screening. A more refined risk-based approach could concentrate elastography and fibrosis evaluation on patients with the highest pretest probability of disease, improving diagnostic yield and reducing low-value testing.
For health systems and policymakers, the study offers a potential pathway to make liver case-finding more sustainable. If validated prospectively and adapted to local epidemiology, refined criteria could help prioritize populations for noninvasive fibrosis assessment, specialty referral, and preventive intervention.
For patients, the findings underscore that liver fibrosis is not simply a binary yes/no condition based on a single risk factor. The overall risk profile matters, and a more personalized approach may be preferable to one-size-fits-all screening.
Conclusion
van Kleef and colleagues provide strong evidence that current fibrosis screening recommendations may be too broad for efficient population use. By refining eligibility to subgroups with at least a 10% risk of elevated liver stiffness, the authors reduced the screened population from roughly 60% to 76% of adults down to about 10% to 22%, while increasing the yield of clinically meaningful abnormalities and preserving strong outcome discrimination.
The study does not settle the optimal national screening policy, but it advances the field toward a more targeted, evidence-based model of liver fibrosis detection—one that may better balance sensitivity, feasibility, and clinical value.
Funding and ClinicalTrials.gov
Funding details were not provided in the abstract. No ClinicalTrials.gov registration was reported in the source citation.
References
1. van Kleef LA, Pustjens J, Schneider CV, et al. Validated early detection metrics reduce the population requiring liver fibrosis screening. Gastroenterology. 2026; published online June 15, 2026. PMID: 42297049.
2. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on non-invasive tests for evaluation of liver disease severity and prognosis. J Hepatol. 2021;75(3):659-689.
3. American Association for the Study of Liver Diseases. Practice guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77(5):1797-1835.
4. Newsome PN, Sasso M, Deeks JJ, et al. FibroScan for non-invasive assessment of liver fibrosis in patients with chronic liver disease: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol. 2019;4(3):218-228.