1.3 Temperature Measurement

At the end of this session you will be able to:

1. Reflect upon and evaluate the most appropriate clinical measurement tool for your own practice setting

In addition you will be able to:

2. Identify and list the core temperature measurement sites and rank them in terms of clinical acceptability for your own practice setting

3. Name and review the key types of temperature measurement instruments and rank them in terms of clinical acceptability for your own practice setting

4. Determine if you are using the most appropriate tools and techniques for temperature measurement in your practice setting

Accurate assessment of a child’s temperature is a crucial element of the care of a child with a fever.

The child’s temperature can be influential in the decisions made about the need for investigations, whether to administer medication or even whether to admit a child to hospital.

Whilst temperature measurement is an everyday nursing activity, there is a conflicting evidence base regarding which is the best way of ‘taking a child’s temperature’. (For some detailed information about the theory behind taking temperature go to  http://www.graduateresearch.com/thermometry/theory.htm)
(A brief history of thermometers can be either be read or listened to at http://www.uh.edu/engines/epi512.htm )

The supposedly ‘simple’ task of measuring a child’s temperature can be made difficult if the child is restless or uncooperative.  The nurse’s skills in preparing the child for the procedure and engaging the child are vital if the measurement is to be undertaken effectively.

Core temperature measurement can be measured at various sites within the body (see Figure 2). (see also detailed general overview of measurement sites http://www.graduateresearch.com/thermometry/sites.htm)

Each site has advantages and disadvantages. The most common sites used within routine practice are axilla, oral and tympanic membrane.

Safety, convenience and child and parental acceptability are key reasons for the axilla, oral and tympanic sites being the most frequently utilised.

Figure 2: Core Temperature Measurement Sites

Rectal temperature measurement is well established and is generally considered to be the ‘gold standard’ for measurement as it is most likely to accurately predict a true core temperature measured at the pulmonary artery (Craig et al., 2002). 

However, rectal temperature measurement is rarely used in the routine management of fever since it is an invasive procedure and it carries with it the risk of rectal perforation. Its use is often limited to inter-operative and critical care settings and when a very accurate measurement is required (Craig et al, 2002).

There are many different thermometry tools available although they broadly fall into three main groupings (see Table 2).

(see http://www.graduateresearch.com/thermometry/devices.htm for a detailed discussion of thermometry devices)

Each tool has specific advantages and disadvantages, with some more suitable for the home setting than the clinical setting. In addition, to their reliability and accuracy, their acceptability to the child, parent and the nurse needs to be considered when choosing which tool to utilise in a given situation or setting. 

Common to all methods of temperature measurement are the requirements that the nurse is accurate, appropriately trained and consistent in the technique, the use of the specific instrument and in documenting the result.

(For useful tips on taking ‘better temperatures’ see http://www.graduateresearch.com/thermometry/tips.htm )

There has been a proliferation of studies in recent years which have investigated the accuracy of the different thermometers and attempted to determine which instrument(s) are the best ones for detecting fever in children.  The results are somewhat contradictory.

The only consensus is that rectal mercury-in-glass thermometers provide the most accurate measurements (Wilshaw et al., 1999) albeit that it is not suitable for general use within clinical practice. Kocoglu et al. (2001) propose that an “optimal thermometer should have the following features: accurate temperature measure; ease of application in a short while; safety and absence of potential risks; and tolerability by the patient” (p39 in the PDF download).

Craig et al’s (2002) systematic review of 44 studies compared infrared ear thermometry with rectal thermometry and concluded that, whilst the mean differences between rectal and ear temperatures were small, there were wide limits of agreement.

They conclude that “infrared ear thermometry does not show sufficient agreement with an established method of temperature measurement to be used in situations where body temperature needs to be measured with precision” (p603).  

This is supported by findings from other studies (Androkites et al., 1998) such as Lanham et al. (1999) who concluded that “sensitivity, specificity, positive predictive value, and negative predictive value are unacceptably low and the number of children with fever who would be missed by screening with  a tympanic thermometer is unacceptable” (p619).

These findings are also supported by Banitalebi & Bangstad (2002) who question the sensitivity of infrared tympanic thermometry and propose that some children with fever are likely to be missed. Jean-Mary et al., (2002) suggest that infrared thermometers have limited accuracy, particularly as the child’s age and rectal temperature increase, but that these thermometers could be used for general screening purposes but not when a clinically accurate measurement is needed.

Other studies such as Barton et al’s (2003) evaluation of the tympanic thermometer with electrical and chemical dot thermometers showed no statistical differences (but suggested that the clinical significance of the results could be debated as clinical dot thermometers placed in the axilla read higher than electronic thermometers). Erickson (1996) proposes that clinical dot thermometers provide “temperature approximations” and that confirmatory readings from “electronic or mercury thermometers with established accuracy are advised”.

Morely et al.’s (1998) study reviewed the accuracy of mercury, Tempa-DOT and FeverScan thermometers and determined that Tempa-DOT had a better predictive value than FeverScan for detecting fever and that FeverScan “seriously over-diagnoses fever by 74%” (p178).

However, despite the discrepancies and inconsistencies noted in relation to electronic, tympanic membrane and clinical dot thermometers compared to the gold-standards they remain clinically useful due to their convenience and acceptability by clinicians, children and parents.

(For further literature see http://www.graduateresearch.com/thermometry/literature.htm)