When African Americans eat less saturated fat (like butter and fatty meat) and more carbs (like fruits and whole grains), their bad cholesterol (LDL) goes down, but another dangerous blood fat called Lp(a) goes up—so one risk goes down while another goes up.
Scientific Claim
Reducing dietary saturated fat intake from 16% to 6% of total energy and replacing it with carbohydrates causes a 10% reduction in LDL cholesterol (−12 mg/dL) and a 24% increase in lipoprotein(a) (Lp(a)) (＋11 mg/dL) in African Americans aged 18–65 with average baseline LDL-C of 116 mg/dL, indicating opposing effects on two key atherogenic lipoproteins.
Original Statement
“Compared to the AAD diet, LDL-C was reduced by the DASH-type diet (mean change: −12 mg/dl)... In contrast, Lp(a) levels increased following the DASH-type diet compared with AAD (median: 58 vs. 44 mg/dl, P < 0.0001).”
Evidence Quality Assessment
Claim Status
appropriately stated
Study Design Support
Design supports claim
Appropriate Language Strength
definitive
Can make definitive causal claims
Assessment Explanation
The study is a well-controlled RCT with randomization, crossover design, and direct measurement of biomarkers under feeding conditions, allowing definitive causal language. The reported effect sizes and statistical significance support precise, definitive verbs.
Gold Standard Evidence Needed
According to GRADE and EBM methodology, here is what ideal scientific evidence would look like to definitively prove or disprove this specific claim, ordered from strongest to weakest evidence.
Systematic Review & Meta-AnalysisLevel 1aWhether the LDL-C reduction and Lp(a) increase observed in this trial consistently occur across diverse African American populations and translate to differential cardiovascular event rates.
Whether the LDL-C reduction and Lp(a) increase observed in this trial consistently occur across diverse African American populations and translate to differential cardiovascular event rates.
What This Would Prove
Whether the LDL-C reduction and Lp(a) increase observed in this trial consistently occur across diverse African American populations and translate to differential cardiovascular event rates.
Ideal Study Design
A meta-analysis of 10+ randomized controlled trials (n≥5000 total) in African Americans comparing low-SFA (≤6%) vs. average-SFA (≥12%) diets, with central lab-measured LDL-C and Lp(a) at 4–8 weeks, and long-term CVD outcomes (MI, stroke, death) tracked for ≥5 years.
Limitation: Cannot establish biological mechanism or isolate effects of specific carbohydrate sources.
Randomized Controlled TrialLevel 1bIn EvidenceWhether replacing saturated fat with unsaturated fats (instead of carbs) blunts the Lp(a) rise while preserving LDL-C lowering in African Americans.
Whether replacing saturated fat with unsaturated fats (instead of carbs) blunts the Lp(a) rise while preserving LDL-C lowering in African Americans.
What This Would Prove
Whether replacing saturated fat with unsaturated fats (instead of carbs) blunts the Lp(a) rise while preserving LDL-C lowering in African Americans.
Ideal Study Design
A 3-arm RCT of 200 African Americans aged 25–60 with baseline Lp(a) ≥30 mg/dL, randomized to: (1) low-SFA/carb-rich diet (6% SFA, 57% carb), (2) low-SFA/MUFA-rich diet (6% SFA, 25% MUFA), (3) control (16% SFA), for 12 weeks, measuring LDL-C, Lp(a), oxidized phospholipids, and apo(a) isoforms.
Limitation: Cannot prove long-term clinical outcomes or genetic interactions.
Prospective Cohort StudyLevel 2bWhether individuals with the largest Lp(a) increases after SFA reduction have higher future cardiovascular events despite LDL-C lowering.
Whether individuals with the largest Lp(a) increases after SFA reduction have higher future cardiovascular events despite LDL-C lowering.
What This Would Prove
Whether individuals with the largest Lp(a) increases after SFA reduction have higher future cardiovascular events despite LDL-C lowering.
Ideal Study Design
A prospective cohort of 3000 African Americans with baseline Lp(a) measured, followed for 10+ years, tracking dietary patterns (via FFQ), serial Lp(a) and LDL-C changes, and adjudicated CVD events, stratified by magnitude of Lp(a) response to SFA reduction.
Limitation: Cannot prove causation due to potential residual confounding.
Case-Control StudyLevel 3bWhether genetic variants in apo(a) size or other loci predict the magnitude of Lp(a) increase in response to low-SFA diets.
Whether genetic variants in apo(a) size or other loci predict the magnitude of Lp(a) increase in response to low-SFA diets.
What This Would Prove
Whether genetic variants in apo(a) size or other loci predict the magnitude of Lp(a) increase in response to low-SFA diets.
Ideal Study Design
A case-control study comparing 200 African Americans with extreme Lp(a) increases (>20 mg/dL) vs. minimal increases (<5 mg/dL) after 8 weeks of low-SFA diet, genotyped for 50+ SNPs in LPA, APOE, and related lipid pathways.
Limitation: Cannot determine if genetic effects are independent of diet or vice versa.
Evidence from Studies
Supporting (1)
Reducing saturated fat intake lowers LDL-C but increases Lp(a) levels in African Americans: the GET-READI feeding trial
This study found that when African Americans ate less saturated fat (like butter and fatty meat) and more carbs, their 'bad' cholesterol (LDL) went down by 12 points, but another risky blood fat called Lp(a) went up—exactly what the claim says.