Why the food database matters more than you think
The accuracy of a meal plan depends entirely on the accuracy of the food data behind it. Use the wrong database for Indian foods and your calorie calculations can be off by 15–25%. Most nutrition software uses USDA FoodData Central as its primary (or only) database. For Indian dietitians, this creates a systematic problem.
What is IFCT 2017?
The Indian Food Composition Table (IFCT 2017) is published by the National Institute of Nutrition (NIN), Hyderabad. It covers 528 commonly consumed Indian foods with nutrient values measured from samples collected across India. The key difference from USDA: IFCT values are based on Indian preparation methods, Indian varieties of ingredients, and Indian portion sizes. IFCT 2017 provides data for 151 nutrients per food.
What is INDB?
The Indian Nutritional Database (INDB) extends IFCT coverage to foods not included in the 2017 edition. It's particularly useful for regional foods, processed foods common in India, and newer food products.
Key differences: IFCT vs USDA for common Indian foods
| Food (per 100g) | IFCT Calories | USDA Calories | Difference |
|---|---|---|---|
| Paneer (Cottage Cheese) | 265 kcal | 321 kcal | +21% |
| Whole Wheat Roti (cooked) | 297 kcal | 264 kcal | -11% |
| Moong Dal (cooked) | 104 kcal | 118 kcal | +13% |
| Poha (Flattened Rice, cooked) | 158 kcal | Not available | N/A |
| Idli (steamed) | 58 kcal | Not available | N/A |
| Ghee | 900 kcal | 876 kcal | -3% |
The paneer discrepancy is particularly significant — a client eating 100g of paneer daily would have their calorie intake underestimated by 56 kcal/day using USDA data. Over a month, that's 1,680 kcal unaccounted for.
The paneer problem, quantified
Using USDA instead of IFCT for paneer overestimates calories by 56 kcal per 100g serving. For a client eating paneer daily, that's 1,680 kcal/month of error — nearly an entire day's intake. This single food can be the difference between a client losing weight and plateauing.
When to use USDA
USDA FoodData Central is still the best choice for several categories:
- Packaged and processed foods — USDA has extensive data on branded products, many now available in India (protein bars, cereals, supplements)
- International foods — clients who eat Western foods regularly (pasta, bread, cheese, breakfast cereals)
- Foods not in IFCT or INDB — USDA's 300,000+ entries fill gaps for uncommon ingredients
- Supplements and fortified foods — USDA has better coverage of fortified products and dietary supplements
- Restaurant chains — USDA includes data from major restaurant chains that operate in India
The right approach: use all three
The most accurate approach for Indian dietitians is to use IFCT as the primary source for Indian foods, INDB for regional and newer Indian foods, and USDA for everything else. This is exactly how MealStack's food database is structured — all three databases are combined into a single searchable interface, with IFCT/INDB results prioritised for Indian foods.
Practical recommendations
- Always use IFCT values for traditional Indian foods (dal, sabzi, roti, rice, idli, dosa, poha, upma)
- Use INDB for regional specialties and newer Indian foods not in IFCT 2017
- Use USDA for packaged foods, international foods, and anything not in IFCT/INDB
- When a food appears in multiple databases, prefer the one with Indian preparation context
- For custom foods, use AI nutrient estimation as a starting point and adjust based on your knowledge
Browse the full database: MealStack Indian Food Nutrition Database — 9,349 foods from IFCT, INDB, and USDA, free to browse.