So I’ve had a Fitbit Charge HR (pictured) for two weeks now and I thought it might interest some people to read how I’m getting on with it.
I should start by saying that before getting it I had read many of the reports, good and bad, on this relatively new line of gadgets that aim to inspire their users to live a healthier life-style. I wasn’t expecting it to be able to calculate exactly how many calories I’d burned each day, precisely how far I’d walked, count meticulously how many floors I’d climbed or improve my sleeping habits. On the other hand, although it was now measuring the pulse from my wrist I was expecting the same accuracy on heart rate that I got from a chest-strap (a running accessory I’d be very happy never to have to wear again!)
With those relatively low expectations, I programmed in my details, charged it up and strapped it to my wrist two weeks ago tomorrow. I have worn it pretty much 24/7 except for when it has needed charging (see my comments later on battery life) and when I’ve been in the shower (although the instructions claim that it can be worn there). I’ve done 11 exercise sessions whilst wearing it including one Parkrun, a number of road runs, runs on a treadmill and exercise bike ‘rides’ in the gym. Despite reservations about wearing it in bed (I don’t wear a watch to bed) I’ve managed that too allowing it to monitor my sleep patterns. I’m reasonably happy therefore to claim that I’ve given it a decent ‘road test’.
I have been successfully monitoring my calorie intake for nearly a year now using http://www.myfitnesspal.com so I was suitably pleased to see that there was a link to it from Fitbit’s website. In its defence, Fitbit does allow you to input what you eat directly into the smart phone application or onto the website itself but I was happy to continue using Myfitnesspal as it already has 99% of the foodstuffs and meals I eat logged into it. So I have continued to enter my meals and snacks into Myfitnesspal and shortly afterwards the number of calories duly registers on my Fitbit app and the website dashboard. I can’t comment on how good Fitbit’s own calorie counting would be. I would say, for those reading this in the UK, that it doesn’t offer a UK based food database when you register.
Myfitnesspal estimates a daily calorie allowance by asking you to assess your daily activity level (sedentary, lightly active, active and very active are the options) then it adds a suitable number of calories to your Basic Metabolic Rate or BMR (the amount of calories you would use if you were to lay in bed all day) plus any daily exercise you enter into it.
Fitbit, of course, is more sophisticated than this. It takes your BMR (in my case about 1,600 calories a day or 67 calories an hour) and adds calories throughout the day depending on how much activity you do (a complex algorithm that seems to take into account your number of steps, floors climbed and heart rate). You can ask your Fitbit to monitor exercise but if, like me, you already have apps that do that (I use Endomondo or Garmin Connect) there are links to their websites too. Once Fitbit receives the data from a session recorded elsewhere it automatically adjusts any calories it has logged during the session to prevent double-counting.
Obviously I don’t know how many calories I burn each day so I have no idea which method has supplied the closest estimate. What I can say however is that come the end of each day, when I have wanted to know whether I have enough calories left for the inevitable snack or glass of wine before bedtime, both methods have been within 10% of each other in telling me how many calories I have left. As Myfitnesspal has helped me to lose nearly 18 kilos over the last year, 100% through calorie counting, I think Fitbit does just as good a job of estimating my calorie needs. The principle difference between the two is that Myfitnesspal works on an ‘average day’ and Fitbit knows whether you’ve sat around all day or been up and about on your feet.
I have absolutely no idea how a small digital wristband measures the number of steps you take. I have taken the time though, at fairly random intervals during the last two weeks, to check my Fitbit before setting off on a walk then counting how many steps I’ve taken. I have found it to be +\- 5% accurate on this. I should mention here that by default your Fitbit sets you a target of 10,000 steps a day (or 8k) and you can set it to vibrate when you achieve one of these goals. I have found this part of the two weeks the most compelling statistic as I’ve found myself regularly checking either the band itself or my phone app, once it has synchronised, to see how I’m progressing towards my 10,000 step goal.
With two green flashing sensors on its underbelly, the Fitbit monitors your heart rate all the while you wear it. I say this without being 100% certain of this. Sometimes when I access the smart phone app (I have an iPhone 5s) and it shows as ‘Connected’ to the wristband, it doesn’t always report a heart rate immediately. I don’t know if this is a software related delay or whether the Fitbit sometimes loses my pulse. I’m not overly worried by this as it seems to monitor it consistently enough to give decent averages and a ‘daily resting heart rate’. This resting heart rate is not to be confused with your minimum heart rate which presumably you reach whilst asleep. This figure seems to be a value, calculated by the monitor, that represents your average heart rate when at rest. Research on various websites reveals that this value should be somewhere between 60 and 80 beats per minute.
This function is the one that intrigues me the most. I don’t sleep very soundly anymore and the adverts for the Fitbit claimed that it would monitor my sleep patterns informing me of when I’d been restless during the night and when I’d actually woken up.
So has it lived up to these claims? I’d say for the most part that it has. I usually go to bed at least half an hour before attempting to get to sleep and the wristband seems to know this. It records a reasonable approximation of when I climb into bed (within 5 minutes or so) and then seems to show me as ‘restless’ until I my head actually ‘hits the pillow’. It only seems to record me as ‘awake’ if I take a trip to the bathroom during the night but it does highlight periods where I’ve stirred – as best I can remember anyway when I see the chart the next day.
The paperwork that comes with the Fitbit claims that the battery lasts for up to 5 days. I have to say that I’ve never had more than 3 days use out of it yet but I have been checking the display rather more frequently than I will once the novelty wears off. I’ve also been regularly running the smart phone app which seems to force a synchronisation between wristband and phone that may also be draining the battery more than the makers estimated the average user would do. In a later blog I may update this assessment as rechargeable batteries, in my experience, are fickle at the best of times and performance may improve as if I let it run down fully once or twice before recharging it.
As I’ve hinted at under other titles I’m impressed by the smart phone app and the website. I can’t comment on how the app works on mobiles other than iPhones but the link between the wristband and my mobile works well. I’ve been viewing the website and editing my dashboard mostly in Chrome on a Windows PC. The same website can be viewed on my iPad (also in Chrome) but it’s not quite as user-friendly as it is on my Windows PC.
As you may have gathered I’m rather please with my new toy. I shall continue to wear it 24/7 (or for as close to that as the battery allows!) I may eventually use it to count my calorie intake and record my exercise but for the moment anyway I’ll stick with the other apps and websites.