Back to evidence libraryPHC Evidence

RCTs Table Comparing Low-Carb to Low-Fat Diets (67 RCTs; 36 favoured low-carb, 0 favoured low-fat)

PHC Summary2019
Key finding

PHC's running compilation of randomised controlled trials comparing low-carbohydrate diets (under 130 g carbs/day) with low-fat diets (under 35% of calories from fat) finds that low-carb consistently outperforms low-fat for weight loss. PHC frames this body of evidence as 'gold-standard' cause-and-effect data that contradicts epidemiology-based guidelines still dominant in NHS and NICE advice.

Supporting data

What the paper found

  • 01

    67 RCTs in the 2019 snapshot (71 in the currently-maintained PHC version).

  • 02

    36 of 67 RCTs significantly favoured low-carb diets; 0 of 67 significantly favoured low-fat.

  • 03

    Across all trials: 3,030 participants on low-carb arms vs 2,062 on low-fat arms.

  • 04

    Trial durations ranged from 30 days to 2 years.

  • 05

    Companion resources: a meta-analyses summary, a printable table, and an infographic visualisation.

Full summary

About this paper

This PHC summary is a living evidence table — not a one-off review — that aggregates every randomised controlled trial the charity has been able to identify comparing a low-carbohydrate diet (under 130 g of carbs per day) with a low-fat diet (under 35% of calories from fat). The version published in February 2019 catalogued 67 RCTs; PHC has continued to add trials since. Each row links to the underlying DOI.

PHC introduces the table with a methodological framing: 'Currently, a lot of nutrition science is based on observational data through epidemiology, which presents association but not causation. Randomised controlled trials are the gold standard in evidence because they help establish cause and effect.' The table's purpose is to let clinicians and the public judge the causal evidence directly.

The headline tally is stark. When only statistically significant results are counted, 36 trials significantly favoured low-carb and zero significantly favoured low-fat. Across all trials, 3,030 participants received low-carb interventions versus 2,062 on low-fat.

PHC notes the trade-off between trial length (30 days up to 2 years) and robustness: longer trials are more expensive and harder to complete but more authoritative, while shorter ones may not give participants time to adapt. The compilation therefore presents results at multiple durations rather than collapsing into a single point estimate.

For clinicians, the practical argument is that UK NICE weight-loss guidance — built around a 600 kcal/day low-fat deficit and unchanged in its fundamentals since 2006 — sits at odds with this body of evidence. PHC's explicit recommendation is that low-carb, higher-fat eating should be added to NHS guidance as an acceptable, effective and safe option.

The table is distributed alongside a printable PDF, an infographic and a companion document on meta-analyses, making it easy for GPs, nurses and patients to consult during clinical conversations.