Low-carbohydrate diet score subtypes and all-cause mortality in general and chronic disease populations: a systematic review and meta-analysis of prospective cohort studies.

OBJECTIVES: To examine associations of overall, healthy and unhealthy low-carbohydrate diet (LCD) scores with all-cause mortality in general and chronic disease populations. Healthy and unhealthy subtypes were compared with assess whether the observed associations depend on macronutrient quality rather than carbohydrate restriction alone. DESIGN: Systematic review and pairwise category meta-analysis. DATA SOURCES: PubMed, MEDLINE, ProQuest Medical Database and Web of Science Core Collection were searched from inception to 12 May 2026. ELIGIBILITY CRITERIA: Prospective cohort studies of adults assessing LCD adherence using a validated three-macronutrient composite score and reporting HRs for all-cause mortality were eligible. DATA EXTRACTION AND SYNTHESIS: Two reviewers independently extracted data and assessed study quality using the Newcastle-Ottawa Scale (NOS). Random-effects meta-analyses compared each higher reported LCD category with the lowest category, stratified by LCD score subtype and population type. Certainty of evidence was assessed using NutriGrade. RESULTS: 18 prospective cohort studies included 779 158 participants and 219 457 deaths; all scored 7-9/9 on the NOS. In chronic disease populations, the highest healthy LCD category was associated with lower mortality than the lowest category (HR 0.72, 95% CI 0.69 to 0.76; I²=0%; high certainty), as was the highest overall LCD category (HR 0.85, 95% CI 0.75 to 0.96; I²=68%; high certainty). In the general population, the highest healthy LCD category was not associated with lower mortality than the lowest category (HR 0.93, 95% CI 0.85 to 1.01; I²=69%; moderate certainty), and neither was the highest overall LCD category (HR 0.96, 95% CI 0.90 to 1.03; I²=87%; low certainty). Unhealthy LCD scores were not associated with mortality in either population. CONCLUSIONS: Healthy LCD adherence was associated with lower all-cause mortality, particularly among individuals with chronic diseases. Unhealthy LCD scores were not associated with mortality in either population, suggesting that macronutrient quality and source may matter more than carbohydrate reduction alone.