Skip to content   Skip to footer navigation 

How accurate is your personal thermometer?

Our investigation into personal thermometers has found you can't always trust the number on the screen.

infrared thermometer on pink background
Last updated: 28 September 2020
Fact-checked

Fact-checked

Checked for accuracy by our qualified fact-checkers and verifiers. Find out more about fact-checking at CHOICE.

Being able to identify when you or your child is running a fever is important. It can help you decide whether or not to see a doctor, and inform how you manage an illness. 

A fever is usually defined as an elevated body temperature of 38°C or higher (normal body temperature is generally 36–37°C).  Since there can be as little as 1°C difference between a healthy temperature and a fever, getting an accurate reading is important. 

But how do you know that your thermometer is giving you an accurate temperature reading?

taking a temperature with infrared thermometer

How accurate are personal thermometers? 

Our personal thermometer reviews test digital probe thermometers, infrared thermometers (forehead and ear) and the Nurofen Feversmart Temperature Monitor which sticks directly to the skin. 

Most thermometers claim to be accurate within 0.1–0.3°C. But our team of experts found that some personal thermometers can be off by as much as 0.83°C, meaning that a healthy temperature of say, 37.4°C could be misread as a fever of 38.2°C, causing unnecessary alarm and even unwarranted trips to the emergency room. 

While we mostly found thermometers which showed incorrectly elevated temperatures, there have also been consumer reports of thermometers showing readings lower than the actual temperature, leading to missed fevers.

Some personal thermometers can be off by as much as 0.83°C, meaning that a healthy temperature of say, 37.4°C could be misread as a fever of 38.2°C, causing unnecessary alarm 

Six of the 15 thermometers in our latest test did not meet their own accuracy claims. All the models that failed to meet their accuracy claims were forehead infrared thermometers (one was a forehead/ear thermometer), as well as the Nurofen Feversmart Temperature Monitor, which is a patch you place directly on your skin. 

We also tested repeatability, which measures the consistency of the thermometer's readings. Six of the 15 thermometers did not achieve good repeatability, meaning they displayed variable temperatures when testing the same person over a short period of time. Those with poor repeatability were all either ear or forehead infrared thermometers.

Which thermometers can you trust?

We recommend thermometers that score at least 75% overall, and are accurate and easy to use, with no unsecured button batteries. Our top rated thermometers give accurate and consistent results and if used correctly you can be confident they are correctly diagnosing your child's temperature. 

A good thermometer is generally considered to be accurate to within 0.3°C. Our top-rated digital probe thermometers can meet their claimed accuracy of within 0.1°C. 

Ear and forehead thermometers are generally less accurate, but our testing has found that the better models are still accurate to within 0.2°C.

Remember to check for other symptoms

Emergency paediatrician Dr Ruth Barker says that while noticing a fever is helpful, monitoring other symptoms is just as important.

"Parents need to remember that the degree of fever is not the sole indicator of illness," says Barker. "It should form part of a broader assessment which involves the child's level of alertness, their colouring – if they look flushed, or pale or grey, their food and water intake and the presence of other symptoms like vomiting.

"Regardless of what the thermometer says, if you think your child is sick, you should seek medical attention."

Case study: A close call with a newborn

Katie* believes she missed a fever in her newborn son because of an inaccurate forehead thermometer.

"One day when my son was about two months old, I noticed he was really irritable and I suspected he might have a fever, but he had no other symptoms," she says. 

Katie took her son's temperature with a forehead thermometer and it came up as 36°C.

"I had a feeling something wasn't right so I tried to book him into the doctor. The receptionist asked if he had a fever and I said no, so she said it wasn't urgent and booked me an appointment the next day.

If only we had identified his fever earlier he could have had Panadol and antibiotics before things escalated

"The next day, before leaving for the doctor, I took my son's temperature again and got the exact same reading as the day before. But when the doctor took his temperature, it showed a very high fever, so my son was immediately given Panadol and taken in the ambulance to the emergency department. 

"We were sent to another hospital where our son was admitted and treated with regular antibiotics and Panadol for his high temperatures. Luckily he recovered and we were able to take him home after two long nights in hospital. We never found out the cause of his illness – it was recorded as an infection of unknown cause.

"I kept thinking that if only we had identified his fever earlier he could have had Panadol and antibiotics before things escalated."

*Not her real name

Case study: A false positive

Jaime says her newborn daughter was admitted to hospital unnecessarily after an inaccurate forehead thermometer reading.

"When my daughter was 11 days old we had the doctor come for a home visit because she was sneezy and snotty," Jaime says. 

The doctor used a thermal forehead thermometer and got a temperature reading of 38.4. He told Jaime her young daughter likely had an infection and needed to head to children's emergency just to be safe. 

"When we arrived, the medical staff said even though she didn't have any symptoms, they needed to check for everything because of the high temperature reading she had had at home," says Jaime. "So they used a cannula to draw bloods and they wanted to take spinal fluids too, but I refused. 

I was so shaken and exhausted by this experience

"We had to stay in hospital for 24 hours before they told us that she had no infection, no symptoms and no fever readings – there was nothing wrong with her at all. 

"I was so shaken and exhausted by this experience. As a first-time mum I trusted the thermometer and the doctor's comments over my own instincts and I still beat myself up over it today."

Case study: Problems in childcare centres

Childcare worker Sarah says she's noticed the forehead scanning thermometers used in her centre are often inaccurate.

"Recently I noticed that one of the babies seemed to have a fever, so I checked and got a reading of 38.0," says Sarah. "It was quite a young baby so I immediately alerted my boss to the high temperature. She came over to check herself and got a reading of 36.7, so said there was no need to alert the parents."

Sarah was certain the baby was unwell and the centre continued to check her temperature throughout the day. 

"We kept getting such different readings that we never ended up alerting the parents," she says. 

"There have been lots of times I'm sure a child has a fever but the inconsistent readings mean that the parents aren't alerted.

We're asking parents to trust us with their kids, but we don't trust the thermometers we're using and the results we are getting

On the flip side, Sarah says there have also been times when the centre has sent kids home with a fever, only for their parents to later report they didn't have a fever at all when they got home. 

"One parent actually brought their own digital probe thermometer from home so that we could get accurate readings for their child.

"It's hard because we're asking parents to trust us with their kids, but we don't trust the thermometers we're using and the results we are getting."

taking the temperature of young girl

How to get an accurate reading

Even if you have a highly accurate thermometer, you can still get an inaccurate reading if you don't perform the test properly. Temperature readings are affected by a number of different factors, including:

Measurement location

Different parts of the body are slightly different temperatures, so you'll get different readings depending on where you test. Ear readings will generally be 0.3 to 0.6°C higher than oral readings, while armpit and forehead readings can be 0.3 to 0.6°C lower than oral readings.

Tip: Make sure you always test at the same location so you can accurately track changes over time. 

Baseline temperature

Different people will have a different healthy baseline temperature, so it's helpful to refer to a known baseline temperature to accurately diagnose a fever.

But body temperature can vary by as much as 0.5°C throughout the day, with lower temperatures in the morning and higher temperatures in the evening. A woman's temperature can also fluctuate with their menstrual cycle – generally it's 0.5°C higher after ovulation, then goes back down when you get your period.

Tip: Take a reading in the morning and the evening when you're feeling well to establish your average healthy body temperature.

Food/drinks and activity

Consuming hot or cold food or drinks can affect oral temperature readings, while exercise or a hot shower can elevate forehead temperature.

Tip: Stay as still as possible during the reading and don't take a forehead reading directly after a bath, shower or exercise. Wait 15 to 30 minutes after eating or drinking to take an oral temperature.

Incorrect use

You need to read the thermometer's instructions and follow them carefully to get an accurate reading. 

Tip: For oral readings, place the probe under the tongue and close the mouth. For armpit readings, keep the arm tightly pressed against the body. For ear readings, make sure the probe is positioned correctly.

Paediatrician's tip: Dr Barker says that many parents simply feel their child's forehead. "I often ask parents if their child feels warm, hot or burning," says Barker. "It's subjective, but the more parents practice it, the easier it gets. It can also be a helpful second check if what the thermometer is telling you doesn't seem right."

Advice for choosing a thermometer 

While choosing a thermometer with accurate readings is important, there are other factors to consider when deciding which model to buy.

Decide which type you want

There are three main types of personal thermometer on the market and each has their advantages and disadvantages.

Digital probe thermometers

These are usually fairly accurate and affordable but some take quite a long time to deliver a reading so may be tricky to use on kids. Dr Barker recommends buying a digital probe thermometer as the most accurate and convenient option that will suit the whole family.

Ear thermometers

These deliver fast readings but can be difficult to position correctly and readings can be affected by ear-wax build up. Dr Barker says parents of babies under 12 months should steer clear of ear thermometers, saying they are tricky to use with small ear canals and therefore can often be out by as much as 2°C.

Forehead thermometers

These are quick and non-invasive, but results can vary depending on skin type, skin colour, sweat, or whether the subject is wearing make-up. They are also not recommended for use on infants under three months.

Features to look for

  • a large, backlit display 
  • an audible beep that tells you when the reading is ready
  • a colour change or tone when a fever is detected
  • a memory function that displays the last temperature reading (or multiple previous readings) recorded so you can track changes
  • a model that doesn't use button batteries or has a safe battery compartment which does not allow children to access dangerous button batteries.

How are thermometers regulated?

A digital thermometer which is intended to be used for diagnosing fevers or monitoring human body temperature is considered a medical device and must be included on the Australian Register of Therapeutic Goods (ARTG). 

This means personal thermometers are required by law to comply with regulatory requirements set out by the Therapeutic Goods Administration (TGA). 

The regulations state that medical devices like thermometers that have a measuring function must provide "accurate, precise and stable measurements" appropriate for that device.  

There is a standard (ISO 80601-2-56:2009) which specifies the general and technical requirements for thermometers, including a requirement that the thermometer is accurate within 0.3°C. 

However this standard is not mandatory to meet TGA requirements, meaning thermometers aren't strictly required to achieve this accuracy in order to be listed on the ARTG.

And while manufacturers are required to undergo auditing by an independent third party and to provide evidence that their product complies with the regulatory requirements, products are not usually tested by the TGA before being sold. 

40% of thermometers in our latest tests didn't meet their own accuracy claims

CHOICE is helping consumers during the COVID-19 crisis by making our investigations and test results on thermometers and hand sanitisers freely available to everyone who needs them.

You can support our independent journalism and testing by chipping in.

Donate

We care about accuracy. See something that's not quite right in this article? Let us know or read more about fact checking at CHOICE.