Study analysis · The British Journal of Nutrition · 2021
Carbs vs. Protein: The 120,000-Person Study That Could Change Your Diet
People who ate more carbs had a higher chance of dying early, while those who ate more protein had a lower chance, but balance is key.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study is like a big survey that follows many people over time, asking them about what they eat and then keeping track of who gets sick. It can tell us if eating certain foods is linked to health problems, but it can't prove that the food actually causes the problem because people who eat one thing might also do other different things that affect their health.
What’s the bottom line?
Researchers looked at what thousands of people ate and followed them for about 11 years to see who got sick or died. They found that eating lots of carbs (like bread and pasta) may be linked to a higher chance of dying, while eating more protein (like meat, fish, eggs) may be linked to a lower chance. Eating a moderate amount of food overall seemed good for heart and brain health.
How strong is this study?
The study is quite good because it had many participants and watched them for many years, while collecting lots of information. But it relies on people remembering what they ate, which might not be perfect, and the people who joined might be healthier than average, so the results might not apply to everyone.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
38 / 100
- Randomizationnot randomized
- Blindingnot blinded
- Control groupno control group
- Sample size (n=120963)+20/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 552 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational cohort study. The study design cannot control for unmeasured confounding variables, and dietary intake is self-reported, which may introduce recall bias and measurement error. Therefore, the observed associations cannot be interpreted as causal.
No Conflicts
No conflicts of interest identified
No conflicts of interest declared; no funding statement present in the provided text. The study appears to be an independent analysis of UK Biobank data.
The provided text is incomplete and lacks any conflict of interest or funding declarations. Therefore, the assessment is based on the absence of disclosed conflicts. The study is observational and based on a large prospective cohort, which typically has lower risk of industry bias. However, without full text, this remains an assumption.
Key takeaways
- 01
People who ate the most carbs had a 13% higher risk of dying during the study compared to those who ate the least.
- 02
People who ate the most protein had an 18% lower risk.
- 03
Moderate total food intake was linked to 13% lower risk of heart disease and 29% lower risk of dementia.
- 04
High sugar intake was linked to a 14% higher risk of heart disease.
- 05
The effects are small-to-moderate.
- 06
For an individual, the change in risk is not huge, but on a population level, it could matter.
- 07
The study finds links, not proof that one food causes these outcomes.
Surprising findings
- High carbohydrate intake was associated with increased mortality, but only 20% of the UK Biobank participants completed the diet recalls, potentially introducing selection bias.We expect large studies to have reliable data, but if only a small, healthier subset completed recalls, results might not apply to the general population.
- Moderate energy intake was linked to lower dementia risk, but this association was not seen for high or low intakes.We often think more food leads to worse outcomes, but here moderation was beneficial, and low intake was not protective.
- Sex differences emerged: high sugar intake was associated with increased mortality in men (HR 1.17) but not in women, and high saturated fat was linked to increased dementia risk in women (HR 1.69) but not in men.Men and women may metabolize nutrients differently, leading to different health outcomes. This challenges one-size-fits-all dietary advice.
Practical takeaways
Balance your plate with moderate calories, prioritize protein (e.g., lean meats, beans, tofu) and limit refined carbohydrates and added sugars.
This is an observational study, so it cannot prove causation. Also, self-reported diet data may be inaccurate, and the study population may not represent everyone.
medium confidenceSwap some carbs for protein and healthy fats; for example, replace a sugary snack with nuts or Greek yogurt.
Don't go to extremes; the study showed moderate energy intake was best, so avoid severe calorie restriction.
medium confidenceIf you're a man, pay extra attention to sugar intake, as it was linked to higher mortality in men specifically.
Sex differences need replication; this could be due to chance or residual confounding.
low confidenceWhy this study matters
Carbs Linked to Higher Death Risk
In a UK Biobank study of 120,963 adults, those in the highest third of carbohydrate intake (as % of energy) had a 13% higher risk of all-cause mortality during 11.1 years of follow-up (HR 1.13, 95% CI 1.03-1.23). This suggests that high-carb diets might be associated with shorter lifespans.
Many people follow high-carb, low-fat diets based on older guidelines. This finding challenges that approach and could make viewers reconsider their macronutrient balance.
Protein: The Protective Macronutrient
High protein intake was associated with an 18% lower risk of all-cause mortality (HR 0.82, 95% CI 0.76-0.89) when comparing the highest third to the lowest. This effect was seen across the whole cohort, suggesting protein may have protective benefits.
Protein is often emphasized for muscle building, but this study shows it might also be linked to longevity. This could encourage viewers to prioritize protein-rich foods.
Moderate Energy Intake Protects Heart and Brain
Compared to low energy intake, moderate total energy intake was associated with 13% lower risk of CVD (SHR 0.87, 0.79-0.97) and 29% lower risk of dementia (SHR 0.71, 0.52-0.96). This suggests that eating too little might be as harmful as eating too much.
Many people are concerned about overeating, but undereating can also be risky. This finding highlights the importance of finding a balance.
Sugar: A Sweet Path to Heart Disease
High sugar intake (highest third) was associated with a 14% increased risk of CVD (SHR 1.14, 1.03-1.27). With 63% of participants exceeding sugar recommendations, this is a significant public health concern.
Sugar is a hot topic; this study provides more evidence linking it to heart disease. Viewers might be motivated to cut back on added sugar.
The Winning Combo: Low Carb, Low Fat, High Protein
A dietary pattern characterized by low carbohydrate, low fat, and high protein was associated with a 16% lower risk of all-cause mortality (HR 0.84, 0.76-0.93), and for men, a 17% lower risk of CVD (SHR 0.83, 0.71-0.97). This cluster analysis suggests that combination matters more than any single macronutrient.
Instead of focusing on one nutrient, this shows that the overall pattern is key. This could simplify advice: prioritize protein and limit refined carbs and excess fats.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Researchers looked at what thousands of people ate and followed them for about 11 years to see who got sick or died. They found that eating lots of carbs (like bread and pasta) may be linked to a higher chance of dying, while eating more protein (like meat, fish, eggs) may be linked to a lower chance. Eating a moderate amount of food overall seemed good for heart and brain health.
Research results
People who ate the most carbs had a 13% higher risk of dying during the study compared to those who ate the least. People who ate the most protein had an 18% lower risk. Moderate total food intake was linked to 13% lower risk of heart disease and 29% lower risk of dementia. High sugar intake was linked to a 14% higher risk of heart disease.
What this means - more context
The effects are small-to-moderate. For an individual, the change in risk is not huge, but on a population level, it could matter. The study finds links, not proof that one food causes these outcomes.
To investigate the association between individual and combinations of macronutrients with all-cause mortality, cardiovascular disease (CVD), and dementia, and to examine sex differences in these associations.
In a prospective cohort of 120,963 UK Biobank participants (57% women) with at least two 24-hour diet recalls, higher carbohydrate intake as a percentage of energy was associated with an increased risk of all-cause mortality (HR 1.13, 95% CI 1.03-1.23 for highest vs lowest third), while higher protein intake was associated with a reduced risk (HR 0.82, 0.76-0.89). Moderate total energy intake was associated with lower risks of CVD (SHR 0.87, 0.79-0.97) and dementia (SHR 0.71, 0.52-0.96). High sugar intake increased CVD risk (SHR 1.14, 1.03-1.27). A dietary cluster characterized by low carbohydrate, low fat, and high protein was associated with lower all-cause mortality (HR 0.84, 0.76-0.93) and, in men, lower CVD risk (SHR 0.83, 0.71-0.97). Sex differences were noted, but findings were generally modest and required confirmation.
Methods Used
Prospective cohort study using data from the UK Biobank, including 120,963 participants (57% women) who completed two or more web-based 24-hour dietary recalls. Macronutrient intakes were expressed as percentages of total energy and categorized into thirds. Cox proportional hazards models estimated hazard ratios (HR) and sub-distribution HR (SHR) for all-cause mortality, CVD, and dementia, adjusting for confounders and accounting for competing risks. K-means cluster analysis identified dietary patterns. Follow-up was a mean of 11.1 years.
Main Finding
High carbohydrate intake (highest third) was associated with a 13% increased risk of all-cause mortality (HR 1.13, 95% CI 1.03-1.23), while high protein intake was associated with an 18% reduced risk (HR 0.82, 0.76-0.89). Moderate total energy intake was linked to lower CVD (SHR 0.87, 0.79-0.97) and dementia risk (SHR 0.71, 0.52-0.96). High sugar intake increased CVD risk (SHR 1.14, 1.03-1.27). The dietary cluster low carbohydrate, low fat, high protein was associated with lower all-cause mortality (HR 0.84, 0.76-0.93) and, for men, lower CVD risk (SHR 0.83, 0.71-0.97).
Confidence Level
Moderate confidence. Large prospective cohort with validated dietary assessment, but observational design, self-reported diet, potential residual confounding, and only 20% of the UK Biobank completing diet recalls limit certainty.
Study Flags
Red Flags
- •Self-reported dietary intake
- •Only 20% of UK Biobank completed diet recalls
- •Potential residual confounding
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
High carbohydrate intake was associated with increased mortality, but only 20% of the UK Biobank participants completed the diet recalls, potentially introducing selection bias.
We expect large studies to have reliable data, but if only a small, healthier subset completed recalls, results might not apply to the general population.
Practical Takeaways
Balance your plate with moderate calories, prioritize protein (e.g., lean meats, beans, tofu) and limit refined carbohydrates and added sugars.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 552 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study is like a big survey that follows many people over time, asking them about what they eat and then keeping track of who gets sick. It can tell us if eating certain foods is linked to health problems, but it can't prove that the food actually causes the problem because people who eat one thing might also do other different things that affect their health.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Large sample size (over 120,000 participants).
- Prospective design with long follow-up (mean 11.1 years).
- Repeated dietary assessments to better estimate habitual intake.
Weaknesses
- Self-reported dietary data, prone to measurement error and recall bias.
- Observational design, cannot fully exclude residual confounding.
- Potential selection bias due to non-participation and loss to follow-up.
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Researchers looked at what thousands of people ate and followed them for about 11 years to see who got sick or died. They found that eating lots of carbs (like bread and pasta) may be linked to a higher chance of dying, while eating more protein (like meat, fish, eggs) may be linked to a lower chance. Eating a moderate amount of food overall seemed good for heart and brain health.
Research results
People who ate the most carbs had a 13% higher risk of dying during the study compared to those who ate the least. People who ate the most protein had an 18% lower risk. Moderate total food intake was linked to 13% lower risk of heart disease and 29% lower risk of dementia. High sugar intake was linked to a 14% higher risk of heart disease.
What this means - more context
The effects are small-to-moderate. For an individual, the change in risk is not huge, but on a population level, it could matter. The study finds links, not proof that one food causes these outcomes.
To investigate the association between individual and combinations of macronutrients with all-cause mortality, cardiovascular disease (CVD), and dementia, and to examine sex differences in these associations.
In a prospective cohort of 120,963 UK Biobank participants (57% women) with at least two 24-hour diet recalls, higher carbohydrate intake as a percentage of energy was associated with an increased risk of all-cause mortality (HR 1.13, 95% CI 1.03-1.23 for highest vs lowest third), while higher protein intake was associated with a reduced risk (HR 0.82, 0.76-0.89). Moderate total energy intake was associated with lower risks of CVD (SHR 0.87, 0.79-0.97) and dementia (SHR 0.71, 0.52-0.96). High sugar intake increased CVD risk (SHR 1.14, 1.03-1.27). A dietary cluster characterized by low carbohydrate, low fat, and high protein was associated with lower all-cause mortality (HR 0.84, 0.76-0.93) and, in men, lower CVD risk (SHR 0.83, 0.71-0.97). Sex differences were noted, but findings were generally modest and required confirmation.
Methods Used
Prospective cohort study using data from the UK Biobank, including 120,963 participants (57% women) who completed two or more web-based 24-hour dietary recalls. Macronutrient intakes were expressed as percentages of total energy and categorized into thirds. Cox proportional hazards models estimated hazard ratios (HR) and sub-distribution HR (SHR) for all-cause mortality, CVD, and dementia, adjusting for confounders and accounting for competing risks. K-means cluster analysis identified dietary patterns. Follow-up was a mean of 11.1 years.
Main Finding
High carbohydrate intake (highest third) was associated with a 13% increased risk of all-cause mortality (HR 1.13, 95% CI 1.03-1.23), while high protein intake was associated with an 18% reduced risk (HR 0.82, 0.76-0.89). Moderate total energy intake was linked to lower CVD (SHR 0.87, 0.79-0.97) and dementia risk (SHR 0.71, 0.52-0.96). High sugar intake increased CVD risk (SHR 1.14, 1.03-1.27). The dietary cluster low carbohydrate, low fat, high protein was associated with lower all-cause mortality (HR 0.84, 0.76-0.93) and, for men, lower CVD risk (SHR 0.83, 0.71-0.97).
Confidence Level
Moderate confidence. Large prospective cohort with validated dietary assessment, but observational design, self-reported diet, potential residual confounding, and only 20% of the UK Biobank completing diet recalls limit certainty.
Study Flags
Red Flags
- •Self-reported dietary intake
- •Only 20% of UK Biobank completed diet recalls
- •Potential residual confounding
No biological mechanisms were identified in this study. This may be an epidemiological, observational, or survey-based study that reports associations rather than proposing causal biological pathways.
Surprising Findings
High carbohydrate intake was associated with increased mortality, but only 20% of the UK Biobank participants completed the diet recalls, potentially introducing selection bias.
We expect large studies to have reliable data, but if only a small, healthier subset completed recalls, results might not apply to the general population.
Practical Takeaways
Balance your plate with moderate calories, prioritize protein (e.g., lean meats, beans, tofu) and limit refined carbohydrates and added sugars.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 552 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
Human Cohort Study
Subject
Moderate probability
on the GRADE evidence scale
This study is like a big survey that follows many people over time, asking them about what they eat and then keeping track of who gets sick. It can tell us if eating certain foods is linked to health problems, but it can't prove that the food actually causes the problem because people who eat one thing might also do other different things that affect their health.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Large sample size (over 120,000 participants).
- Prospective design with long follow-up (mean 11.1 years).
- Repeated dietary assessments to better estimate habitual intake.
Weaknesses
- Self-reported dietary data, prone to measurement error and recall bias.
- Observational design, cannot fully exclude residual confounding.
- Potential selection bias due to non-participation and loss to follow-up.
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
The study is quite good because it had many participants and watched them for many years, while collecting lots of information. But it relies on people remembering what they ate, which might not be perfect, and the people who joined might be healthier than average, so the results might not apply to everyone.
0 / 100
- COI disclosureconflicts of interest not disclosed
- Data availabilitydata not shared
- Code availabilitycode not shared
38 / 100
- Randomizationnot randomized
- Blindingnot blinded
- Control groupno control group
- Sample size (n=120963)+20/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registrationnot pre-registered
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 552 / 100
Probability of being correct
Groups of people are followed over time to see who develops an outcome. Strong for identifying risk factors and associations, but cannot prove causation as firmly as RCTs.
This design cannot establish causation — the findings describe an association, not a cause. This is an observational cohort study. The study design cannot control for unmeasured confounding variables, and dietary intake is self-reported, which may introduce recall bias and measurement error. Therefore, the observed associations cannot be interpreted as causal.
No Conflicts
No conflicts of interest identified
No conflicts of interest declared; no funding statement present in the provided text. The study appears to be an independent analysis of UK Biobank data.
The provided text is incomplete and lacks any conflict of interest or funding declarations. Therefore, the assessment is based on the absence of disclosed conflicts. The study is observational and based on a large prospective cohort, which typically has lower risk of industry bias. However, without full text, this remains an assumption.