Health and fitness apps, such as MyFitnessPal, have exploded in popularity over the past few years. In 2015, there were already 165,000 commercially available. Heart rate measuring apps have proven particularly popular, with some simply requiring the user to point their phone camera at their face.

According to Dr Christophe Wyss, a cardiologist at Heart Clinic Zurich and author of the study, it is “human nature” to want to try them out, even if you are feeling healthy.

“The problem is that there is no law requiring validation of these apps and therefore no way for consumers to know if the results are accurate,” he said.

Some heart rate apps work by touching the fingertip to the phone’s camera (contact photoplethysmography), while others work by simply holding the camera in front of the user’s face (non-contact photoplethysmography). Photoplethysmography works by illuminating the skin and measuring changes in how much light is absorbed as the heart pumps blood around the body.

The study, published in the European Journal of Preventive Cardiology, tested four commercially available heart rate apps on more than 100 patients. Two of these apps used the contact method and two used the non-contact method. Results were compared with those taken from an ECG – the clinical gold standard for heart rate measurement, which measures the electrical activity of the heart – and fingertip pulse oximetry, a technique based on photoplethysmography.

The researchers found that there were significant differences between the ECG measurement and the app measurement: these differences varied by up to 20 beats per minute.

The contact apps performed better than non-contact apps, particularly at higher heart rates and lower body temperatures. Under these conditions, the non-contact apps tended to overestimate heart rate.

Perhaps most surprising was the difference the researchers found between readings from the two contact apps. This variation arose in spite of the apps being based on the same technology. The variation could not be explained by variations with camera technology, age, body temperature or heart rate itself.

“The one contact app was excellent, performing almost like a medically approved pulse oximeter device,” said Dr Wyss. “But the other app was not accurate, even though they use the same technology.”

“The difference in performance between the contact apps is probably down to the algorithm the app uses to calculate heart rate which is commercially confidential,” said Dr Wyss. “It means that just because the underlying technology works in one app doesn’t mean it works in another one and we can’t assume that all contact heart rate apps are accurate.”

“Consumers and interpreting physicians need to be aware that the differences between apps are huge and there are no criteria to assess them,” he added.

“We also don’t know what happens to the heart rate data and whether it is stored somewhere, which could be an issue for data protection.”

Despite their popularity, health and fitness apps have come under criticism for not having proven positive effects. A 2016 University of Pittsburgh study found that fitness apps may not help users lose weight, and NHS-endorsed mental health apps have been shown to be overwhelmingly ineffective.