Geriatric Nutritional Risk Index (GNRI) Calculator

Geriatric Nutritional Risk Index (GNRI) Calculator

Calculate the Geriatric Nutritional Risk Index from serum albumin, weight and height, with the original risk bands and what the score is really measuring.

Geriatric Nutritional Risk Index (GNRI)

Albumin + weight ratio → GNRI
An albumin reported in g/L must be divided by 10 first — 34 g/L is 3.4 g/dL.
Ideal body weight is taken as the weight giving a BMI of 22.
90.0GNRIExample

Serum albumin 3.4 g/dL, weight 60 kg, height 170 cm

Bouillanne's formula

GNRI = (14.89 × serum albumin in g/dL) + (41.7 × present weight ÷ ideal weight)
Ideal body weight = 22 × height in metres²
The weight ratio is capped at 1
albumin
in g/dL — the equivalent formula with albumin in g/L uses a coefficient of 1.489
ideal body weight
the weight giving a BMI of 22 kg/m², which is the definition used in the original paper
cap at 1
when present weight exceeds ideal weight the ratio is set to 1, so obesity cannot inflate the score

Worked example

Serum albumin 3.4 g/dL, weight 60 kg, height 170 cm
Ideal body weight = 22 × 1.70² = 63.58 kg
Weight ratio = 60 ÷ 63.58 = 0.944 (below 1, so not capped)
Albumin term: 14.89 × 3.4 = 50.6
Weight term: 41.7 × 0.944 = 39.4
50.6 + 39.4 = 90.0 → moderate nutritional risk

Risk categories

GNRICategory
> 98No nutrition-related risk
92 – 98Low risk
82 – 91Moderate risk
< 82Major risk
The bands are those described with the original index. They classify nutrition-related risk of morbidity and mortality, not nutritional status itself.

How much each input moves the score

ChangeEffect on GNRI
Albumin falls by 0.5 g/dL (5 g/L)−7.4 points
Albumin falls by 1.0 g/dL (10 g/L)−14.9 points
Weight falls by 5 kg at 170 cm−3.3 points
Weight rises above ideal body weightNo effect — the ratio is capped at 1
Albumin dominates: a modest fall in albumin outweighs a substantial loss of weight, which is why inflammation has such a large influence on the result.

What the GNRI is measuring, and why albumin dominates it

The Geriatric Nutritional Risk Index is 14.89 times the serum albumin in g/dL plus 41.7 times the ratio of present body weight to ideal body weight, where ideal body weight is the weight giving a BMI of 22 kg/m². The weight ratio is capped at 1, so a patient heavier than their ideal weight scores the same on that term as one exactly at it, and obesity cannot inflate the result. If the laboratory reports albumin in g/L, the equivalent formula uses a coefficient of 1.489 — or divide the g/L figure by ten before entering it here.

It is a nutrition-related risk index, not a measure of nutritional status, and the distinction is not pedantic. Bouillanne and colleagues derived it in 2005 in hospitalised elderly patients as a modification of the Nutritional Risk Index, replacing usual weight — which older inpatients often cannot recall reliably — with ideal weight calculated from height. What it predicts is the risk of complications and death associated with nutritional state, which is not the same as answering whether a given patient is malnourished.

Albumin dominates the score arithmetically, and albumin is a negative acute-phase reactant. Inflammation, infection, liver disease and nephrotic syndrome all depress it independently of nutrition, and capillary leak and fluid shifts move it further. A low GNRI in an acutely unwell patient may therefore be measuring inflammation rather than nutrition. Reading the score alongside a CRP is the practical safeguard, and a score that recovers as the CRP falls was never describing intake in the first place.

That entanglement is also why the index performs so well. Because it captures inflammatory burden as well as body mass, it predicts mortality and complications across a very wide range of settings — haemodialysis, heart failure, cancer, surgery and critical care — and it has been adopted well beyond the elderly medical inpatients it was built for. Use it to stratify risk and to trigger a formal assessment, not to decide whether a patient needs feeding: that decision rests on a structured assessment such as the GLIM criteria or the Subjective Global Assessment.

Frequently asked questions

How is the GNRI calculated?

GNRI = 14.89 × serum albumin in g/dL + 41.7 × (present weight ÷ ideal weight), with ideal weight taken as the weight giving a BMI of 22 kg/m². With an albumin of 3.4 g/dL, a weight of 60 kg and a height of 170 cm, the GNRI is 90.0.

What are the GNRI risk categories?

Above 98 is no nutrition-related risk, 92 to 98 is low risk, 82 to 91 is moderate risk, and below 82 is major risk. They describe nutrition-related risk of morbidity and mortality rather than nutritional status.

My albumin is reported in g/L — what do I enter?

Divide by 10, so 34 g/L is 3.4 g/dL. The equivalent published formula for albumin in g/L uses a coefficient of 1.489 instead of 14.89, which gives the same result.

What happens if the patient weighs more than their ideal weight?

The weight ratio is capped at 1, so it contributes a maximum of 41.7 points however heavy the patient is. Obesity therefore cannot inflate the score, and the result then depends entirely on albumin.

Does a low GNRI mean the patient is malnourished?

Not necessarily. Albumin dominates the score and is a negative acute-phase reactant, so inflammation, infection, liver disease and nephrotic syndrome all depress it independently of intake. Interpret a low score alongside a CRP and a structured nutritional assessment.

Related calculators

References

  1. Bouillanne O, Morineau G, Dupont C, et al. Geriatric Nutritional Risk Index: a new index for evaluating at-risk elderly medical patients. Am J Clin Nutr. 2005;82(4):777–783.
  2. Cederholm T, Jensen GL, Correia MITD, et al. GLIM criteria for the diagnosis of malnutrition — a consensus report from the global clinical nutrition community. Clin Nutr. 2019;38(1):1–9.
  3. Volkert D, Beck AM, Cederholm T, et al. ESPEN practical guideline: clinical nutrition and hydration in geriatrics. Clin Nutr. 2022;41(4):958–989.