Reference audit · US/Canada and Europe · Checked November 20, 2026
A water target is useless until you know what it counts. A recommendation for total water cannot be pasted into a plain-water bottle tracker without changing its meaning. Water in other drinks and food has not vanished because the app only has a button for glasses.
This is why the familiar argument about whether everyone needs two, three or four liters often starts in the wrong place. Before debating the number, inspect the definition. We compared two reference systems and built a small accounting worksheet that makes the missing categories visible. It is a reference review, not an assessment of anyone's hydration.
Two reference systems, one easily missed word
The US and Canadian Dietary Reference Intakes give adult Adequate Intakes of 3.7 L/day for men and 2.7 L/day for women. Those values include drinking water, other beverages and food moisture. Adequate Intake is a population reference used when evidence is insufficient to establish an Estimated Average Requirement; it is not the same as an RDA.
EFSA's European reference gives 2.5 L/day for adult men and 2.0 L/day for adult women, also including all those sources. Its stated conditions are moderate environmental temperature and moderate physical activity. These are different reference frameworks, not two precise prescriptions that can be averaged into a better one.
| Framework | Men | Women | Quantity counted |
|---|---|---|---|
| US/Canada DRI | 3.7 L/day | 2.7 L/day | Total water |
| EFSA | 2.5 L/day | 2.0 L/day | Total water |
Neither row means that every adult should finish that volume of plain water in addition to meals and other drinks. Nor does falling below a reference value on a diary establish dehydration. A target is not a diagnostic test, and a log is only as complete as its entries.
A worked day that a bottle counter misreads
Consider a completely hypothetical day with 1.50 L of plain water, 0.75 L of water supplied by other drinks and 0.65 L of water supplied by food. Total water is 2.90 L. A bottle-only log shows 1.50 L. Both numbers can be arithmetically correct while describing different things.
| Field | Water amount | Running total |
|---|---|---|
| Plain water | 1.50 L | 1.50 L |
| Water in other beverages | 0.75 L | 2.25 L |
| Food moisture | 0.65 L | 2.90 L |
Against a hypothetical total-water target of 2.70 L, comparing only the bottle field creates an apparent 1.20 L gap. Comparing the full invented day gives 0.20 L above that reference instead. That does not tell the person to drink less or more. It shows that the conclusion changed because the accounting boundary changed.
The worksheet uses water content, not the gross weight of food. A 300 g meal does not automatically contribute 300 mL of water. Likewise, a drink's volume and its water content need not be identical. If food moisture is unknown, leave it unknown. Filling the blank with zero silently turns an incomplete diary into an apparently complete one.
Do not turn a population average into your missing-data patch
The DRI report describes food supplying roughly one-fifth of total water in the underlying population data. That is useful context, not permission to add exactly 20% to every person's bottle log. The fraction depends on what was eaten and drunk. A generic allowance should be labeled as an estimate and must not be presented as observed intake.
There is a second trap here. If an app already includes a meal's water contribution, adding a separate blanket food allowance counts some of it twice. The solution is to identify the source of each contribution, not to choose whichever total makes the progress ring look complete.
What a useful hydration log should say
- Name the target. Is it plain water, all beverages, or total water? Keep the source and population beside it.
- Name each input. Distinguish a measured drink from an estimated food-water contribution.
- Keep unknowns visible. Missing food data is not evidence that food supplied none.
- Do not diagnose from the percentage. A completed target is not proof of hydration, and an incomplete target is not proof of dehydration.
Exercise, heat, illness and individualized fluid restrictions make a universal bottle target especially unhelpful. This page does not provide a sweat-replacement protocol or override a clinician's fluid advice. It addresses a simpler mistake that should be fixed before personalization begins: comparing different quantities under the same label.
Method and reusable worksheet
We manually extracted the adult total-water references and their scope, then constructed the example above. We did not measure intake, sweat losses or hydration biomarkers. Download the CSV or JSON with source dates and assumptions. The numerical example is Vora's original teaching model, not a participant record.
Publisher disclosure: we make Vora, which brings proactive AI together with nutrition, workouts and wearable context. Our interest is in making health records interpretable. The electrolyte dilution reference addresses a separate question: how much sodium a prepared volume contains. Neither worksheet requires the app. Corrections: support@askvora.com.