This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

Gold, incentives and meh

Dehydration in children - diagnosis by numbers

Dr Nick Summerton shows how numerical data derived from the history and examination can help in diagnosing dehydration in children

Dr Nick Summerton shows how numerical data derived from the history and examination can help in diagnosing dehydration in children

Dehydration is often a concern when dealing with children suffering vomiting, diarrhoea or a poor oral fluid intake.

41199984Although I have been unable to identify any studies from primary care, much research has been undertaken in hospital and emergency department settings. The likelihood ratios for these are detailed in table 1.

The problem with many of these signs is that they are highly non-specific and vary considerably between observers. The kappa values are detailed in table 2.


Likelihood ratio

This is the ratio of the probability of an event (such as a symptom) in diseased persons to the probability of that same event in non-diseased persons.

Likelihood ratios indicate how many times more (or less) likely a result is in a patient with the disease compared with a patient free of disease.

The adjectives ‘positive' or ‘negative' indicate whether the likelihood ratio refers to the presence of the clinical information (positive) or the absence of the clinical information (negative).

• Positive likelihood ratios with the highest value argue most for disease when the clinical information is present.

• Negative likelihood ratios with the value closest to zero argue the most against disease when that information is absent.

Kappa values

Kappa values measure reliability – the extent to which repeated measurements of a stable phenomenon produce similar results. This level of agreement is expressed as a chance-corrected proportional value.

Kappa values greater than 0.80 may be taken to represent almost perfect agreement, values between 0.61 and 0.80 substantial agreement, values between

0.41 and 0.60 moderate agreement, values between 0.21 and 0.40 slight agreement and values of 0.20 or below represent poor agreement beyond chance.

From this evidence, it seems that sunken eyes, a sunken fontanelle and dry mucous membranes offer little help in making a positive diagnosis of 5% dehydration because of the low likelihood ratios and the range of kappa values (from poor to moderate agreement).

Dehydration is less likely in the absence of dry mucous membranes as this exhibits the lowest negative likelihood ratio and also has a more acceptable range of kappa values then many of the other physical signs.

However, in assessing children it is always important not to ignore the information elicited from the parents.

The parental report of normal urine output (for example, as assessed by the number of wet nappies during a defined time period) has been reported as having a negative likelihood ratio of between 0.16 and 0.27.
Capillary refill time and skin turgor have the highest positive likelihood ratios and it may be that much of the inter-observer variation and unreliability may reflect differences in how these signs are elicited. Some advocate a more standardised approach to enhance reliability (see clinical assessment box below).

Dr Nick Summerton is a GP in Yorkshire and clinical lead at BMI Healthcare

This article is based on Patient-Centred Diagnosis by Dr Nick Summerton, published by Radcliffe Publishing, ISBN 9781846190551

Pulse readers can claim a 15% discount when they buy direct from Radcliffe by telephoning 01235 528820 or online at by putting PULSEPCD in the discount code box at the checkout

Baby Box 1. Likelihood ratio for 5% 2. Kappa ranges for symptoms Key messages Clinical assessment

How to assess capillary refill time

The room should be well-lit and warm. The arm should be held at the level of the heart and pressure should be gradually increased on the palmar surface of one of the patient's fingertips until the capillary bed blanches. The pressure is then released immediately and, in non-dehydrated children, the colour should return in less than two seconds.

How to assess skin turgor

The examiner should use the thumb and index finger to pinch a small skin fold on the lateral abdominal wall at the level of the umbilicus. The fold is then promptly released and the time taken to return to normal is classified (somewhat unsatisfactorily) as immediate, slightly delayed or prolonged.

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say