Data Pipeline
1. Primary Source: Harvard T.H. Chan SPH Oxalate Database (2024)
Our oxalate values come from direct laboratory measurements using ion chromatography — the gold standard for food oxalate analysis. The database was compiled at the Harvard T.H. Chan School of Public Health, analyzed by Dr. John Knight's lab at the University of Alabama at Birmingham, and published through the Oxalosis & Hyperoxaluria Foundation (OHF).
These are point measurements from specific samples, not database averages, not crowd-sourced estimates, not derived calculations. Each value represents one food sample analyzed in one lab. This is simultaneously the database's greatest strength (precise) and its limitation (single-sample variability).
Harvard T.H. Chan School of Public Health / OHF, 2024.
Access: Harvard Oxalate Database
2. Nutrition Matching: USDA FoodData Central
Each Harvard food entry is manually matched to USDA FoodData Central (SR Legacy 2018 and Foundation Foods) to retrieve sodium, protein, calcium, and water content. Foods with confirmed USDA matches are designated Tier 1 (complete); foods with oxalate only are Tier 2 (oxalate only). Every Tier 1 food carries a traceable USDA FDC ID in our source database.
3. Decision Model: Three-Color Verdict
Our rating system applies clinically grounded oxalate thresholds to each food's per-serving value:
- 🟢 Safe: Under 25 mg oxalate per serving — recommended for daily consumption in normal portions.
- 🟡 Caution: 25–99 mg oxalate per serving — manageable with portion control and calcium pairing.
- 🔴 Avoid: 100 mg or more per serving — limit or substitute for kidney stone patients.
Importantly: sodium, animal protein, and calcium richness are displayed as independent informational tags — they do not affect the color verdict. This is intentional. Sodium is a stone risk factor, but a separate mechanism from oxalate. Animal protein is an AUA-flagged risk factor, but a separate mechanism. Calcium from food is protective. We show all four dimensions (oxalate + sodium + protein + calcium) independently so users can make multi-factor decisions.
4. Static Site Generation
All pages are pre-generated as static HTML directly from our food database JSON — no backend, no runtime database queries, no JavaScript-rendered content. This ensures fast load times (inline CSS, zero external dependencies), reliable indexing by search engines, and no dependency on third-party APIs.
Clinical Alignment
Our dietary guidance is grounded in peer-reviewed literature and clinical practice guidelines. Eight core references are cited on every food page:
Taylor EN, Curhan GC. "Oxalate intake and the risk for nephrolithiasis." J Am Soc Nephrol 2007;18(7):2198-2204.
PMID: 17538185 — Harvard cohort (HPFS + NHS), n=240,681. Found 22% increased stone risk in highest vs lowest oxalate quintile.
Borghi L et al. "Comparison of two diets for the prevention of recurrent stones." N Engl J Med 2002;346(2):77-84.
DOI: 10.1056/NEJMoa010369 — Normal calcium + low animal protein + low salt reduced recurrence from 38% to 20% vs low calcium diet.
Pearle MS et al. "Medical Management of Kidney Stones: AUA Guideline." J Urol 2014;192(2):316-324.
DOI: 10.1016/j.juro.2014.05.006 — Official AUA guideline: fluid intake, dietary calcium, sodium restriction, moderate animal protein.
Ferraro PM et al. "Dietary and lifestyle factors for primary prevention of nephrolithiasis." BMC Nephrol 2020;21(1):267.
PMID: 32652950 — DASH-style diet reduced stone risk by 31%; fluid, coffee, tea, beer had inverse associations.
Fink HA et al. "Diet, fluid, or supplements for secondary prevention of nephrolithiasis." Eur Urol 2009;56(1):72-80.
DOI: 10.1016/j.eururo.2009.03.031 — High fluid intake reduced recurrence (RR 0.39); soft drink reduction helped high consumers.
Liebman M, Costa G. "Effects of calcium and magnesium on urinary oxalate excretion." J Urol 2000;163(5):1565-1569.
PMID: 10751889 — Calcium co-ingestion reduced urinary oxalate by 25-50% by binding oxalate in the gut.
Holmes RP, Assimos DG. "The impact of dietary oxalate on kidney stone formation." Urol Res 2004;32(5):311-316.
DOI: 10.1007/s00240-004-0437-3 — Only ~10-15% of population absorbs meaningful dietary oxalate; gut Oxalobacter formigenes is the key variable.
Lin BB et al. "Dietary treatment and fluid intake for prevention of recurrent calcium stones." PLoS One 2021;16(4):e0250257.
PMID: 33872340 — Low sodium + normal calcium diet reduced recurrence; water intake was the single most effective intervention.
Data Quality Commitments
- No estimates. We do not average across sources. We do not impute missing values. The published Harvard/OHF number is shown as-is.
- Full traceability. Every oxalate value is traceable to its source data file. Every Tier 1 nutrition value is traceable to a USDA FDC ID.
- Transparent tiers. Tier 1 (Harvard + USDA) and Tier 2 (Harvard only) are clearly labeled on every food page.
- Independent dimensions. Oxalate, sodium, protein, and calcium are four separate risk dimensions. Our verdict is oxalate-only; the other three are informational. We do not collapse them into a single "score" because that would obscure trade-offs.
Limitations
- Single-sample variability. Oxalate content varies between crop varieties, growing conditions, soil composition, and cooking methods. Harvard's values represent point measurements. Two spinach samples from different farms may differ by 15-25%. We show the published value.
- Individual tolerance varies. Only ~10-15% of the population is a net dietary oxalate absorber (Holmes & Assimos, 2004). The remaining 85-90% excrete most dietary oxalate. 24-hour urine oxalate testing is the only way to know your personal absorption rate.
- Stone type matters. Our verdicts are optimized for calcium oxalate stone formers (~80% of all stones). Uric acid, struvite, and cystine stones have different dietary triggers.
- Cooking effects are food-specific. We cite published cooking-effect ranges (30-50% reduction via boiling), but these are food-dependent. Our "boiled" data points represent specific preparation methods tested by Harvard.
- Not all foods are tested. Our database covers 1,188 foods. This is the largest publicly available oxalate dataset, but it is not exhaustive. Many processed foods, regional specialties, and seasonal items have not been analyzed.
This is dietary reference information. Consult a registered dietitian or urologist for personalized dietary guidance based on your 24-hour urine chemistry.