Do automatic fall alarm solutions really work?

Tens of solutions for automatic detection of falls among elderly are being developed worldwide. But do they really have an effect? This is what we are trying to find out in our trial in Trondheim, Norway. It turns out we are one of very few in the world doing this.

Smart Belt
The “Smart Belt” fall alarm app being used in the trial.

SINTEF SIT and NTNU/St.Olavs Hospital are now involved in an ongoing real world trial of automatic fall alarms in Trondheim. The trial is done in collaboration with Trondheim municipality represented by Laugsand and Byneset health and welfare centers. More than 20 elderly people are involved. The trial will result in one-of-a-kind insights into how this new type of fall alarm technology works. It will also tell us how this technology affects existing alarm services. The trial is funded by the European Commission as part of our ongoing R&D project FARSEEING [1].

Fall alarms belong roughly to one of two types:

  • Manual: Most products in use are in this category. They consist of a button that is pressed manually by the user when a fall happens.
  • Automatic: These are coined as the next generation products. These products automatically generate alarms when a fall happens, without the user needing to do anything. Automatic alarms are useful in situations where the user is not able to manually push a button. This can for instance be an unconscious user, or a user with dementia.

Our trial involves both types and will provide comparison data about the advantages or disadvantages of automatic alarm technology.

Another of many fall alarm technologies being developed or used in FARSEEING

The number of fall alarm products is growing fast. Via our cooperation with Norwegian municipalities we know that service providers are very much affected by this market press. We need scientific data about the effect of these technologies on users and services. We need to know the pitfalls, if any. And we need guidelines on how to improve existing technology and practice. This is what the trial in Trondheim aims at. Our ongoing study of existing literature has unveiled that there are very few scientific publications providing this knowledge. The knowledge we will gain will be useful not only for service providers– developers and vendors of such technology will as well find the results of the trial useful.

The trial will run during autumn 2014. Each user will be observed for a period of one month. If any falls happen during this period, they will be detected by both the existing conventional alarm service and a new automatic alarm. We will then evaluate the sensitivity of the new automatic alarm technology compared to the existing alarm. A good level of sensitivity will mean all real falls are detected and no false falls are detected. We will also collect data about how the introduction of automatic alarms affects the existing practice. For instance, is the level of trust in the technology changed among the elderly wearing these alarms? How does the level of attention to an alarm change among the health personnel when they know it is an automatic and not a manual alarm? How much, if any, time is saved dealing with alarms? The results from the pilot will be published and you will be the first ones to hear about them on this blog!

[1] FARSEEING project web site.
[2] Link to our web page for the literature study (currently not much here since we are in the process of preparing a publication).

Leave a Reply

Your email address will not be published. Required fields are marked *