Telecare Soapbox: Falling outside the box

[Editor Donna’s note: Even if you have already read this great article by Steve, read on to the first comment as to ‘what happened’.]

Before you read the rest of this item, read or re-read the following two popular recent falls-related articles:

Accelerometers, false positives/negatives and fall detection (Tom Doris)

A five-point rebuttal to ‘Accelerometers, false positives/negatives and fall detection’ (Andy Schoonover)

The acceptability of false positive fall alerts and the risk of false negatives has dogged fall detection ever since the first detector was invented. It is great to see this fundamental matter still being debated on this website by experienced practitioners!

However, given that calibration of devices has, as Tom Doris points out, inherent problems however sophisticated they become, I see no one putting their finger on what I consider to be the ‘real’ issue. That is: where in the system is the intelligence that judges whether a fall is problematic or not?

Current systems place the intelligence either with a call centre operator or (as Mike Orton highlights in his comment on Andy Schoonover’s response to Tom) with emergency responders. Andy locates an additional element of intelligence in the decision of someone to accept that they are likely to have more calls from a call centre than are really necessary.

But, what if… what if… the intelligence in the system could be located with the person who has fallen or who just sat down heavily?

An alarm button is a proxy for this but comes with a number of other problems and, in effect, just alerts the system that intelligence from a call centre operator is required.

How would it be if, on detecting what seemed to be a fall, the device spoke to the person and asked if they were okay? And then, if they said they were, took no further action. Or if the device got a negative answer, or none, it would then alert whoever had been assigned to respond? On the other hand, the person could ask the device to call for help at any time they were feeling unwell, fall or not.

By locating the intelligence in the system with the person who has most intelligence about the situation you will have eliminated all false positives and false negatives!

As Tom says in his How to Recognize the Solution When It Arrives conclusion “The next-generation fall detection system will probably need to use more than just accelerometer data to achieve the necessary accuracy and reliability levels.”

Can such a device be developed?

No need!

I flagged up in 2010 that such a device was in development. (Innovative: Game-changing: Next generation (Verity))

In 2013 the developer announced that it was ready for production. Have telecare companies been beating down his door to snap up the rights?

The answer, as you have probably guessed from my tone, is ‘no’ and he has turned his attention to developing a health monitoring product in the belief that there is more interest in the market for health devices. There is now not even a mention of Verity on his company’s current website. iMonSys.

I am fully aware that history is littered with brilliant inventions that didn’t get to market because the developer focused on perfecting the engineering rather than on the marketing needed to ‘cross the chasm’ but it exasperates me that no one in the industry has the wit to invest in taking forward an elegant solution to such a difficult problem.

Steve Hards is founder and ex-editor in chief of TTA. He has no financial interest in the above-mentioned device.

Categories: Latest News and Soapbox.

Comments

  1. Hi Guys

    Good to hear that someone thinks the solution we developed in 2010 would be a good idea.

    We went even further that your proposed solution, and developed artificial intelligent responses to any situation the user found themselves in. From ‘lost’ to ‘going to bed’ to ‘having a bath’ to ‘not wearing the bracelet and the battery is low’ We even embedded the wearers location in the event of an incident so the carer could find them if they went walkabout, eliminating the need for a search party.

    Our Verity system would eliminate 95% of false alarm calls to the call centres. I even had our local MP lobbying for recognition and adoption of Verity to no avail. We have produced numerous academic papers on the ‘state recognition’ capabilities of Verity, but I think my big mistake was ‘it didn’t count calories’ which seems to be more of a driver than saving peoples lives.

    If I had the marketing capability of Nike or Apple, Verity would be on the shelves now, so if there is anyone out there who would like to market Verity, contact me.

    What disappoints me was the reaction we received from the establishment denying that the future of Telecare will be on a completely mobile platform, in order to try and preserve their pensions. Does anyone actually CARE about providing solutions for the elderly and infirm?

    Graham Priestley
    MD iMonSys

  2. The challenge facing designers of all Assisted Living Technologies is understanding that people are heterogeneous in all aspects of their lives from their goals to the issues that prevent them achieving the higher Quality of Life that they want. It is unlikely (if not impossible) therefore to develop a single product that will solve everyone’s problems, and we must instead look for compromises. These will work when individuals are at the heart of the decision process and can make meaningful and informed judgements. This requires a lot more awareness of the options and some lessons in how devices might work in a range of different situations.

    The problem of detecting (and then preventing) falls offers an excellent example of how sensors and intelligent monitoring systems can satisfy users and improve the confidence of some people while making others feel useless because they activate so often. One man’s fall is another man’s stumble, and one woman’s collapse to the ground is another woman’s gradual slip. All could be embarrassing but not a cause of injury, but might also be the start of a downward spiral, especially if the real problem is an inability to get back on one’s feet. So there are uses of language and terminology that we can get wrong in the same was as there are device options that we can fail with, even when we have solved the initial problem of getting someone to accept that technology can help. Companies are very successful at persuading people to want what they don’t need – it’s a shame that they don’t put the same level of effort in getting them to understand that they don’t always want what they do actually need! That remains the role of health, social care and housing professionals along with relatives and friends most of whom will not appreciate the limitations and advantages of a worn device over embedded sensors nor the significance of where it should be worn for maximum sensitivity. I suspect that most don’t have the independently verified information that they need to recommend the best solutions – and with the use of personal budgets increasing, there is a real danger that more people will end up buying sub-optimal solutions, with the worry that this will give the entire Assisted Living Technology industry a bad name.

    Fortunately, this problem is likely to be addressed next year with the launch of Vivo, a new web portal which has been created through collaboration between the Mi Liverpool team (a dallas community project), and T-Cubed, an independent telecare specialist company. Vivo will guide users to find the best assisted living technologies to help people enjoy more independent lives. The unique processes embedded within the website support a novel approach to assessment which focuses on the issues and/or conditions that prevent people from achieving their personal goals for independence. By providing further filters for features and circumstances, it will give users a voyage of product discovery across the whole of the market so that users can make genuine choices without falling into the traps posed by poor levels interoperability and additional dependencies.

Leave a Reply

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