Speaker: Dr. Mona Calvo
This is a summary of a presentation I attended at the National Kidney Foundation Spring Clinical Meeting in New Orleans on May 8, 2026.
Where are P-additives found?
There are 32 functions for food additives, 26 of them include a phosphorus additives e.g. anticaking, stabilizers. Texturizers, etc. This enables several phosphorus additives to be used in a single product.
Phosphorus additives can fall within:
- GRAS (generally recognized as safe) – 49 unique P-additives
- Inorganic – attached to a cation
- E.g. Calcium phosphate
- Organic – attached to a carbon molecule
- Most are modified food starches
- The other type is lecithin
- Inorganic – attached to a cation
- Ok for organic foods – 7 unique P-additives.
- Lecithin is the most widely used phosphate additives – it belongs to 10 functional classes.
The problem with GRAS systems is that there are no limits to the amount that can be added. This is unique to the US and Canada. In Europe GRAS isn’t really used – there are actually specific limits to the amount of these additives that can be added to food.
What are the barriers to determine how much phosphorus is in food?
The Nutrition Facts Tables
In 2016 the nutrition facts tables were updated in the US. This impacted phosphorus labeling because the %DV went from 1000 to 1250mg/day. With a bigger denominator this means that the %DV may appear lower but include more phosphorus.
The Ingredient Lists
Are not clear. Not all phosphorus additives are clearly labeled. Some appears just as modified starches or lecithin making it difficult to determine which products contain additives.
Most modified starches are phosphorus based. In Europe theses are used too but they are labeling more specifically with the specific name rather than the generic “modified starch”.
Another concern is that P-additives can be added in the formulation or processing stage of food preparation and these aren’t required to be on the ingredient list. Only specific ingredients added at the final stage of food processing are required on the label. This is likely a large contributor of why P-content of ultra-processed foods are so poorly estimated.
Under-estimation in Nutrient Databases
25-35% underestimation of phosphorus content will occur if ingredient lists are used to populate nutrient databases compared to nutrient analysis. This was clearly demonstrated but Sullivan et al in 2007 who chemically analyzed chicken products. The minimally processed foods without additives the nutrient databases over-estimated the P content. But foods containing phosphorus additives had P contents that were underestimated.
P-Additives in Plant-Based Alternatives
Most of these products are ultra-processed and contain P additives. Though with increased marketing of plant-based foods people may be preferentially chosing these foods.
Why is it so challenging to make regulatory changes?
We use National Dietary Surveys to estimate P intake but there is a lack of accurate information in the databases that are used to estimate intake. This makes it very challenging to demonstrate causal relationships between P intake and CKD.
What are the links between P-additives and health risk?
Multiple diseases are associated with excess P intake. Excess P intake may cause:
- Oxidative stress
- Inflammation
- Changes in the microbiome
- Endocrine Dysregulation
Increased plasma P increases iPTH levels This stimulates bone osteocytes and FGF23. FGF23 decreases kidney excretion of P and calcitriol activation. This leads to increased calcification in the heart related to hypertension and atherosclerotic plaque.
Insulin resistance can occur related to oxidative stress.
Why do P-additives behave this way?
- Plant derived phosphate is the slowest absorbed. In part related to phytates.
- Animal derived phosphate is faster than plant-based. But still require enzymatic cleaving.
- Additives is highly bioavailable and tends to be quickest absorbed.
Fast absorption leads to increase serum P levels, stimulating iPTH, stimulating FGF23 leading to inhibition of iPTH and decreased kidney synthesis of active form of vitamin D. Leading to calcification.
What evidence do we have that this is happening?
In one study, by Roman-Garcia et al (2010) published in Bone demonstrated this.
Calvo et al in 1990 published in JCEM completed circadian studies on 10 healthy young adults. These adults ate all their meals in the facility. In this 4-week cross-over feeding study feed a low P and high P-additive diet. Subjects were then studied for 24 hours. There was a persistent increase in PTH in a high P/low Ca diets compared to the control arm. There was also a marked decrease in calcitriol synthesis.
In a more recent feeding study, 39 health adults and 11 adults with CKD were given 2 weeks of a high P additive diet followed by 6 weeks of a low additive diet. In this study both urine P decrease, FGF23 concentrations decreased and so did PTH levels.
In a cohort study published in 2023 by Chen et al, food category by UPF was explored for T2DM risk. UPF refined breads, UPF meats, Ready to eat meals, sauses and spreads and artificial sweeteners all lead to increased T2DM risk.
In a study Calvo and Dunford used shopping receipts to determine if the specific foods associated with T2DM also included phosphate additives. Of the top categories of foods sold in the US, 56% of these foods contained additives and these were associated with T2DM risk foods. Suggesting a possible mechanism between phosphorus additives were promoting oxidative stress and inflammation may be the underlying mechanism linking these foods with T2DM risk.
Estimation methods for determining P-additive Intake
The top individual additives were:
- Lecithin
- Sodium phosphates
- Calcium phosphates
- Modified Food Starch (acts as an emulsifier/thickener)
- Sodium Acid Pyrophosphate
In 2014, NHANES data was used to estimate P intake and at that time we knew it under-estimated intake and didn’t account for P-additives. But there was an inflection point at 1400mg P intake and there was an association with increased all cause mortality.
Another method uses a broad level estimation of underestimated P intake by 25% added to other estimates. When this correction factor is applied most people will be well above the EAR for P.
Currently Dr. Calvo is working with comprehensive NHANES data to better estimate P intake and found that in models showing the highest P levels of intake, mortality levels increased. This work hasn’t been published yet.
What she is trying to do is to create overwhelming arguments for changes in policy.
What would be better?
- Establish front of package labeling for the general population to make healthier food choices
- But even better would be to have a kidney specific logo so that it would be easier to support health choices.
- This would be similar to what the American Heart Association has done
