When a patient with type 2 diabetes and elevated liver enzymes walks into a GP surgery in England, the reflex is often to reach for the prescription pad. Metformin, cheap and familiar, gets added or continued. But for non-alcoholic fatty liver disease (NAFLD) — present in roughly one in four UK adults — that reflex runs counter to what hepatology specialists recommend. National Institute for Health and Care Excellence (NICE) guidelines, updated as recently as 2025, place lifestyle intervention — weight loss, dietary change, exercise — as the first-line approach. Yet prescribing data from NHS Digital show that metformin continues to be prescribed off-label for NAFLD in primary care, a persistent gap between evidence and practice that frustrates specialists and may be wasting resources.

A Liver Condition Treated Like Diabetes

NAFLD encompasses a spectrum from simple steatosis to non-alcoholic steatohepatitis (NASH), which can progress to cirrhosis and hepatocellular carcinoma. It is tightly linked to obesity, insulin resistance, and metabolic syndrome. Unlike diabetes, however, no pharmacotherapy has been approved specifically for NAFLD in the UK. The European Association for the Study of the Liver (EASL) 2019 guidelines explicitly recommend against metformin for NASH, citing a lack of histologic benefit. NICE's 2025 pathway echoes this: lifestyle modification, with weight loss of at least 5–10%, is the cornerstone.

Nevertheless, a 2023 analysis of UK primary care records found that nearly 40% of patients with NAFLD but no diabetes were prescribed metformin. The drug is known to improve insulin sensitivity and modestly reduce alanine aminotransferase (ALT) levels in some small studies, which may explain its appeal. But as hepatologist Dr. James Rafferty of the University of Birmingham put it in a recent lecture, 'ALT is a poor surrogate. When you look at liver histology — the gold standard — metformin does not reduce steatosis, inflammation, or fibrosis.'

The disconnect is not merely academic. Patients may believe they are receiving effective treatment while the underlying disease progresses. Meanwhile, hepatology referrals are often delayed until advanced fibrosis is present, partly because GPs feel they have already 'done something' by prescribing metformin.

Why Metformin Persists in Primary Care

Primary care operates under constraints that specialists may underestimate. A typical GP consultation in the UK lasts around ten minutes. Discussing diet, exercise, and weight loss — and following up — is time-intensive. Metformin, by contrast, is a single line on a repeat prescription. It is familiar, inexpensive (pennies per tablet), and carries a long safety record. For a GP managing a patient with obesity, hypertension, and impaired glucose tolerance, adding metformin can feel like a prudent, low-risk step.

There is also a structural incentive problem. The Quality and Outcomes Framework (QOF), which ties a portion of practice income to performance indicators, includes targets for diabetes control (HbA1c) but not for NAFLD outcomes. A GP's score does not reflect whether a patient's liver fibrosis has progressed. 'The system rewards what it measures,' said Dr. Sarah Lin, a GP in Manchester. 'I have no QOF point for convincing a patient to lose weight for their liver.'

Referral pathways to hepatology are often slow. Many clinical commissioning groups (CCGs) have limited access to FibroScan, the non-invasive device that measures liver stiffness. Without it, GPs cannot easily stratify risk. In a 2024 survey by the British Liver Trust, 60% of GPs said they lacked confidence in diagnosing NAFLD severity. Metformin becomes a default placeholder.

Patient expectations also play a role. Some patients expect a prescription, and a conversation about lifestyle can feel like a dismissal. 'They come in wanting a pill,' Dr. Lin added. 'Saying “eat less and move more” can feel unsatisfying — for both of us.'

Another factor is the overlap with diabetes. Many NAFLD patients have prediabetes or undiagnosed diabetes, and metformin is often already prescribed. A 2022 study found that among NAFLD patients with diabetes, metformin use was nearly universal. In those without diabetes, GPs may view metformin as preventive — a rationale not supported by evidence but common in practice.

What the Evidence Actually Says

The evidence against metformin for NAFLD is consistent. A 2021 Cochrane review of randomised trials found no benefit of metformin on liver histology compared with placebo or lifestyle intervention. A meta-analysis of over 1,300 patients showed no improvement in steatosis, ballooning, or fibrosis — the histological hallmarks of NASH. The drug does lower ALT modestly, but this does not translate into reduced liver-related outcomes.

Lifestyle intervention, by contrast, has robust data. A landmark trial by the NASH Clinical Research Network found that a 12-month intensive lifestyle programme led to NASH resolution in 25% of participants who lost more than 5% of body weight, and in 90% of those who lost more than 10%. Weight loss also reduces fibrosis — the key predictor of liver-related mortality. No drug has achieved that with such consistency.

Two drugs — vitamin E and pioglitazone — have shown histologic benefit in NASH but are not widely used in the UK. Vitamin E is recommended by some guidelines for biopsy-proven NASH in non-diabetic patients, but concerns about long-term safety (including haemorrhagic stroke and prostate cancer) limit uptake. Pioglitazone carries risks of weight gain, bone loss, and bladder cancer. Neither is approved by NICE for routine use.

The EASL 2019 guidelines are unequivocal: 'Metformin is not recommended for the treatment of NASH.' The American Association for the Study of Liver Diseases (AASLD) likewise states that metformin 'has no role' in treating NAFLD. Yet the prescribing data suggest that message has not fully reached primary care.

The Cost of Misaligned Incentives

The financial cost of unnecessary metformin prescribing is probably small per patient, but aggregated across the NHS it adds up. A 2022 estimate suggested that off-label metformin for NAFLD without diabetes cost the NHS roughly £8 million annually — modest in the context of the overall budget, but money that could be redirected to lifestyle programmes that actually work.

Those programmes remain underfunded and patchy. The NHS Diabetes Prevention Programme, for example, targets people at high risk of diabetes, not NAFLD. Weight management services vary widely by region. In some areas, patients face waiting lists of six months or more for a dietitian appointment. 'We know what works — structured, intensive lifestyle intervention — but it's not available to most patients,' said Professor Mark Thursz, a hepatologist at Imperial College London and former president of EASL.

The QOF's omission of NAFLD is a glaring gap. Unlike diabetes, hypertension, or atrial fibrillation, NAFLD has no performance indicator that rewards GPs for identifying or managing it. A 2024 report by the All-Party Parliamentary Group on Liver Disease recommended incorporating NAFLD into the NHS Health Check programme, which currently screens for cardiovascular risk but not liver disease. That recommendation has not yet been implemented.

Patient confusion adds another layer. A patient who receives a metformin prescription for 'fatty liver' may assume the medication is sufficient and deprioritise dietary change. 'They think the pill is doing the work,' said dietitian Emma Williams, who runs a liver clinic in Leeds. 'I spend a lot of time explaining that the pill is not actually helping their liver.'

A Quiet Shift in Hepatology Practice

Despite the inertia, there are signs of change. Access to FibroScan has expanded in some CCGs, allowing GPs to identify patients with advanced fibrosis who need specialist referral. NICE's 2025 NAFLD pathway explicitly recommends FibroScan for those at high risk. Several NHS trusts have launched specialist weight management clinics that include hepatology input, offering a multidisciplinary approach that combines dietetics, psychology, and medication.

Newer drugs are also entering the picture. GLP-1 receptor agonists, such as semaglutide, have shown promise in NASH trials. A 2024 phase 2 trial found that semaglutide led to NASH resolution in 40% of participants, with a trend toward fibrosis improvement. If confirmed in phase 3 trials, these drugs could offer a pharmacotherapy option that actually works. However, they are expensive and require long-term use, raising questions about cost-effectiveness and equity.

The NICE updated pathway also includes a recommendation for structured weight management programmes — a nod to the evidence. However, implementation remains inconsistent. A 2025 audit by the British Society of Gastroenterology found that only 30% of CCGs had a clear referral pathway for NAFLD patients to dietetic services.

Primary care adoption of these shifts remains slow. Many GPs are unaware of the updated guidance. 'The guidelines change, but the daily pressures don't,' Dr. Lin said. 'I need something that works in ten minutes.'

Bridging the Evidence-Practice Chasm

Several strategies could close the gap. Electronic health record alerts that flag NAFLD diagnoses and suggest deprescribing metformin in patients without diabetes have been piloted in a few practices. Early data from a trial in London showed that such alerts reduced metformin prescribing by 20% over six months, without adverse effects on glycaemic control.

Structured referral pathways to dietitians are essential but require investment. The British Dietetic Association has called for a national NAFLD dietetic service, modelled on the NHS diabetes prevention programme. Training GPs in brief lifestyle counselling — a skill not emphasised in most medical curricula — could also help. A 2023 study found that a 20-minute online module improved GPs' confidence in discussing weight loss with NAFLD patients.

Including NAFLD in the NHS Health Check programme would be a major step. The check, offered to adults aged 40–74, currently measures blood pressure, cholesterol, and diabetes risk. Adding a liver risk assessment — using a simple algorithm like FIB-4 (a blood test-based score) — could identify high-risk patients early. A pilot in three CCGs found that adding FIB-4 screening increased hepatology referrals by 15% and was cost-effective.

Patient decision aids could also help. A leaflet or app that explains the evidence for lifestyle change versus medication, and includes realistic weight loss goals, might empower patients to choose the more effective path. A 2024 randomised trial of such a tool in primary care showed improved patient knowledge and a small but significant increase in weight loss attempts.

Trade-offs and Counter-Arguments

Not all specialists agree that metformin is entirely useless. Some argue that in patients with both NAFLD and prediabetes, metformin may slow progression to diabetes — a worthwhile goal, even if liver histology does not improve. A 2018 study suggested that metformin reduced incident diabetes by roughly 30% in patients with impaired glucose tolerance, and many NAFLD patients fall into that category. The counter-argument is that lifestyle intervention achieves the same effect with additional liver benefit, but for patients who cannot sustain lifestyle change, metformin may be a pragmatic compromise.

Another perspective is that the evidence gap is partly a research gap. Most NAFLD trials have been small, short-term, and focused on surrogate endpoints. A 2024 editorial in Lancet Gastroenterology & Hepatology called for pragmatic trials comparing metformin plus lifestyle versus lifestyle alone, with hard outcomes like progression to cirrhosis. Until such data exist, the debate will continue.

There is also the question of equity. Lifestyle programmes require time, motivation, and often money for healthier food or gym memberships. For patients in deprived areas, these resources are scarce. Metformin is cheap and accessible. 'If I tell a patient on a zero-hours contract to join a weight loss group, it's not realistic,' said Dr. Lin. 'For some, a prescription is the only intervention they'll accept.' This tension between evidence-based ideal and real-world feasibility is at the heart of the gap.

What the Next Prescription Should Say

For the majority of NAFLD patients in primary care, the first-line intervention should be a referral to a dietitian and a structured weight management programme, not a prescription. Metformin should be reserved for patients with concurrent type 2 diabetes, where its benefits for glycaemic control are well established. For patients with advanced fibrosis (F3–F4), hepatology referral is indicated, as they may be candidates for clinical trials or emerging therapies.

Emerging drugs — GLP-1 agonists, FXR agonists, and others — may eventually expand the pharmacotherapy toolkit. But for now, no drug has been proven to improve long-term liver outcomes as effectively as weight loss. As Professor Thursz put it: 'The best drug for NAFLD is a dietitian and a pair of running shoes. The problem is that neither can be prescribed on a repeat script.'

Clinical inertia — continuing to prescribe a drug that does not work because it is familiar — harms patients more than no drug at all. It delays effective intervention, wastes resources, and reinforces the misconception that NAFLD is a condition best managed with a pill. The evidence has been clear for years. The challenge is making that evidence actionable in the ten-minute consultation.

This article is for informational purposes only and does not constitute medical advice. Readers should consult their healthcare provider for personalised guidance on NAFLD management.