Low-Carb vs. Low-Fat — How Much Should You Cut Back on Starchy Staples? The Answer Lies in “What You Replace Them With,” Not “The Ratio” [2026 Evidence Edition]
[Level 1–2 (High–Moderate)] [Food & Nutrition] [Series: Health for Serious Middle-Agers]
⚠️ Before You Read This Article
This article provides educational information based on peer-reviewed studies (long-term RCTs, meta-analyses, and large cohort studies) on low-carbohydrate and low-fat diets. It is not a substitute for medical advice. If you have type 2 diabetes and are on medication, substantially reducing carbohydrate intake carries a risk of hypoglycemia, so always consult your physician and adjust your medication accordingly.
Key Points of This Article (3-Line Summary)
- For weight loss, low-carb and low-fat deliver the same result at one year (DIETFITS N=609; BMJ 2020 meta-analysis of 121 RCTs). Neither genotype nor insulin secretion could predict which approach “suits” a given person.
- The relationship between carbohydrate ratio and mortality is U-shaped, with the lowest risk at 50–55%. Even on a low-carb diet, replacing carbs with animal-based foods increases mortality (HR 1.18), while replacing them with plant-based foods decreases it (HR 0.82). What you replace them with is everything; the ratio is not.
- In type 2 diabetes, there is a short-term (3–6 month) advantage of HbA1c −0.3 to 0.5%, but the difference disappears at one year. For ketogenic diets (under 50 g carbohydrate), observational data show a doubling of cardiovascular events in people whose LDL spikes; a self-directed approach is not recommended for those in their 50s–60s.
1. Weight Loss: The Same After One Year
Stanford’s DIETFITS (Gardner et al., JAMA 2018, N=609, 12 months) compared low-fat (48% carbohydrate) and low-carb (30%) diets. Weight loss was −5.3 kg vs. −6.0 kg, a difference of 0.7 kg (95% CI −0.2 to 1.6): no significant difference. What drew attention was that neither genotype (P=0.20) nor insulin secretion (P=0.47) could predict “which diet suits whom.”
The network meta-analysis by Ge et al. (BMJ 2020, 121 RCTs, 21,942 participants) likewise found equivalent weight loss at 6 months, 4.63 kg for low-carb vs. 4.37 kg for low-fat, and the difference vanished at 12 months. For lipids, each had a small advantage: LDL −7.1 mg/dL with low-fat, HDL +2.3 with low-carb. Two-year RCTs (Sacks et al., NEJM 2009; Foster et al., 2010) also showed no difference. The exception is the low-carb advantage in Shai et al. (NEJM 2008), but the overall pattern is “convergence at one year.” The main driver of this convergence is declining adherence: in the end, whichever diet you can stick with wins.
The limitations of DIETFITS are that participants were under 50 and non-diabetic, and that both groups shared the instruction to “avoid sugar and refined grains.”
2. Longevity: Carbohydrate Ratio Is U-Shaped, and What You Replace It With Is Everything
Seidelmann et al. (Lancet Public Health 2018, ARIC N=15,428, 25-year follow-up) reported a U-shaped relationship between carbohydrate ratio and mortality, with the lowest risk at 50–55%. In the meta-analysis (N=432,179), HR was 1.20 below 40% and 1.23 above 70%.
What matters decisively is the replacement. Even among low-carb eaters, the group that replaced carbs with animal protein and fat had an HR of 1.18, while the group that replaced them with plant-based foods had an HR of 0.82. The evidence cited for “low-carb diets shorten lifespan” (RR 1.31 in Noto et al. 2013; RR 1.22 in Mazidi et al. 2019) is, in effect, about “people who replaced carbs with meat and animal fat.” Shan et al. (JAMA Intern Med 2020, N=37,233) showed that healthy low-carb and healthy low-fat diets both had an HR of about 0.9, while the unhealthy versions of both had an HR of about 1.07, demonstrating that food quality, not ratio, determines the outcome.
Limitations: all of these are observational studies, dietary assessment relied on FFQs, even the “low-carb group” consumed 35–40% carbohydrate (not a ketogenic diet), and there is confounding from a higher proportion of smokers in the low-carb groups.
3. Type 2 Diabetes: Low-Carb Wins Short-Term, but the Difference Disappears at One Year
The American Diabetes Association (ADA; Evert et al., 2019) explicitly states that “reducing carbohydrate intake has the most evidence for improving glycemia, but there is no optimal ratio.”
- Sainsbury et al. (Diabetes Res Clin Pract 2018, 25 RCTs, 2,412 participants): with carbohydrate below 26%, HbA1c fell by −0.47% at 3 months and −0.36% at 6 months, with no difference at 12 and 24 months. Moderate restriction (26–45%) showed no difference.
- Goldenberg et al. (BMJ 2021, 23 RCTs, 1,357 participants): diabetes remission at 6 months was 57% vs. 31%, nearly disappearing at 12 months.
- Japanese RCT (Sato et al., Clin Nutr 2017, N=66, 130 g carbohydrate/day, 6 months): HbA1c −0.65% vs. 0.00%. Small and short-term.
For people with diabetes or prediabetes, the short-term improvement in blood glucose is real, and 130 g/day (a little over two bowls of rice) is the practical lower limit established in the Japanese RCT. However, since the difference disappears after one year, this should be done under medical supervision with medication adjustment.
4. Safety of Ketogenic Diets (Under 50 g Carbohydrate)
In Iatan et al. (JACC Adv 2024, UK Biobank, 305 people on a low-carb high-fat diet vs. 1,220 controls, 11.8 years), LDL and ApoB rose, and major cardiovascular events were 9.8% vs. 4.3%, HR 2.18. Although observational, the long-term safety of people whose LDL spikes sharply (so-called lean mass hyper-responders) has not been established.
Note that the 1-year KETO-CTA paper (JACC Adv 2025), frequently cited by ketogenic-diet advocates for its claim that “ApoB does not predict plaque,” was retracted in May 2026. A self-directed ketogenic diet is not recommended for those in their 50s–60s.
5. Japanese Data: White Rice and Diabetes, and the Value of Brown Rice
- JPHC (Nanri et al., Am J Clin Nutr 2010, approximately 60,000 people, 5 years): in the highest quartile of white rice intake, OR for diabetes in women was 1.65; in men it was non-significant (with a trend among the physically inactive). Hu et al. (BMJ 2012, meta-analysis) also found RR 1.55 for highest vs. lowest intake in Asians.
- JACC (Eshak et al., Am J Clin Nutr 2014): no association between rice and stroke, heart disease, or cardiovascular death (HR 0.97).
- JPHC (Akter et al., Clin Nutr 2021): U-shaped relationship between low-carb score and all-cause mortality, but plant-based low-carb showed a linear decline in mortality (all-cause HR 0.89, cardiovascular death 0.82).
- Whole grains and dietary fiber (Reynolds et al., Lancet 2019): dietary fiber of 25–29 g/day was associated with a 15–30% reduction in mortality, cardiovascular disease, and type 2 diabetes.
The picture that emerges: white rice “slightly raises diabetes risk but is neutral for cardiovascular health,” while partial substitution with brown rice, barley, and mixed grains is clearly beneficial via dietary fiber.
6. The Pros and Cons of Low-Fat: What WHI Teaches Us
In the WHI Dietary Modification trial (Howard et al., JAMA 2006, 48,835 women aged 50–79, 8.1 years), a low-fat diet yielded HR 0.97 for coronary heart disease and 1.02 for stroke, i.e., no cardiovascular prevention benefit. The weight difference was 1.9 kg at one year and 0.4 kg at 7.5 years. On the other hand, at 19.6 years of follow-up (Chlebowski et al., J Clin Oncol 2020), breast cancer mortality fell to HR 0.79. Low-fat is safe and may reduce breast cancer deaths, but weight loss is small and there is no cardiovascular benefit; a further limitation is that the trial did not distinguish “fat quality.”
7. Conditional Conclusions
| Condition | Guideline |
|---|---|
| Non-diabetic, normal weight | No evidence supports drastically cutting staples. Aim for 50–55% carbohydrate, replace part of your white rice with brown rice, barley, or mixed grains, and get at least 25 g of dietary fiber |
| Obese, non-diabetic | Low-carb and low-fat are equivalent at one year. Choose whichever you can sustain. Cut one bowl of rice (~55 g carbohydrate) 1–2 times per day, and replace it with fish, soy, nuts, and vegetables |
| Type 2 diabetes or prediabetes | Short-term HbA1c advantage of −0.3 to 0.5%. Use 130 g/day as a lower-limit guideline, under medical supervision with medication adjustment |
| Ketogenic diet (<50 g/day) | Observational data show a doubling of cardiovascular events in people whose LDL spikes; no long-term RCTs. Self-directed approach not recommended |
| Everyone | What you replace it with is everything: white rice → animal fat or processed meat is unfavorable; → plant-based protein and fat is favorable |
Framing the question as “carbs or fat?” has little meaning in light of the evidence. The outcome is determined not by what you cut, but by what you replace it with.
8. Frequently Asked Questions
Q. Is there any point in skipping starches at dinner? A. It is effective as a way to reduce total intake, but if you replace it mainly with meat, it works against you. Replace it with fish, soy, and vegetables.
Q. Is brown rice really better than white rice? A. Data point in a favorable direction for diabetes and cardiovascular health through increased dietary fiber. You can expect benefits even from “replacing half” rather than all of it.
Q. Will I lose muscle on a low-carb diet? A. You will with extreme restriction combined with low protein. Follow the guideline from our protein article (1.0–1.2 g/kg) at the same time.
References and Sources
- Gardner CD, et al. (2018). Effect of low-fat vs low-carbohydrate diet on 12-month weight loss (DIETFITS). JAMA 319:667-679. PMID: 29466592
- Ge L, et al. (2020). Comparison of dietary macronutrient patterns of 14 popular named dietary programmes for weight and cardiovascular risk factor reduction. BMJ 369:m696. PMID: 32238384
- Sacks FM, et al. (2009). Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N Engl J Med 360:859-873.
- Seidelmann SB, et al. (2018). Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. Lancet Public Health 3:e419-e428. PMID: 30122560
- Shan Z, et al. (2020). Association of low-carbohydrate and low-fat diets with mortality among US adults. JAMA Intern Med 180:513-523. PMID: 31961383
- Evert AB, et al. (2019). Nutrition therapy for adults with diabetes or prediabetes: a consensus report. Diabetes Care 42:731-754.
- Sainsbury E, et al. (2018). Effect of dietary carbohydrate restriction on glycemic control in adults with diabetes. Diabetes Res Clin Pract 139:239-252. PMID: 29522789
- Goldenberg JZ, et al. (2021). Efficacy and safety of low and very low carbohydrate diets for type 2 diabetes remission. BMJ 372:m4743. PMID: 33441384
- Sato J, et al. (2017). A randomized controlled trial of 130 g/day low-carbohydrate diet in type 2 diabetes with poor glycemic control. Clin Nutr 36:992-1000. PMID: 27472929
- Iatan I, et al. (2024). Long-term low-carbohydrate high-fat diet, LDL-C, and cardiovascular events (UK Biobank). JACC Adv 3:100924. PMID: 39081652
- Nanri A, et al. (2010). Rice intake and type 2 diabetes in Japanese men and women (JPHC). Am J Clin Nutr 92:1468-1477. PMID: 20980490
- Akter S, et al. (2021). Low carbohydrate diet and all cause and cause-specific mortality (JPHC). Clin Nutr 40:2016-2024. PMID: 33046262
- Reynolds A, et al. (2019). Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet 393:434-445. PMID: 30638909
- Howard BV, et al. (2006). Low-fat dietary pattern and risk of cardiovascular disease (WHI). JAMA 295:655-666. PMID: 16467234
Evidence current as of October 2026. KETO-CTA 2025 (JACC Adv) has been retracted and is therefore not cited.
