This is my Graph.

31 Oct 2018
Bookmark

On Wednesday 10th October, I wrote about why it might be a good idea to do your score even when you know you are going to get a score you don't like.

So, here is my graph. Seven years' worth of hard data...

If you're on the free version of Moodscope, then I know you don't have access to your whole history. At the end of this blog, I hope you will see the benefit of spending the price of a couple of cappuccinos a month on getting the full version.

For those of you who have not known me for all those seven years, a quick introduction. You can find it on Friday 24th May 2013, "Hello – I'm Mary and I'm Bipolar", but I joined Moodscope much earlier, on Saturday 14th May 2011.

In those days Jon was still writing all the blogs, and I signed up to Moodscope on the back of a Radio 4 programme. It was my therapist who said, "I've just listened to this, and I think it's just the thing for you." She was right: Moodscope has been my lifeline ever since.

How I wish Moodscope had been available all my life! My first episode of severe (bi-polar) depression occurred when I was seven years old and it has affected me all my life since; although I was only diagnosed at age 43, thirteen years ago. But, just the last seven years' data is useful, as I think you can see.

We don't really know what we are like. In my mind I had a cycle. Every two years I would get depressed. Every four years I would get VERY depressed. At least, that was the pattern I had come up with in the mental health assessment clinic. It was the best I could remember.

But our memory is unreliable at best and a bloody liar at worst.

You can see for the first two and a half years I measured, things weren't too bad. I had some downs, but they were short-lived. The highs were manageable. After January 2014 things started to get worse.

But – without this graph, I wouldn't have known how much worse.

Oh – I was still in denial, until October 2016. That "high" nearly cost me my dearest friend. I am still not sure if we have yet totally recovered. The crash was costly too. My poor family, of course, suffered most.

It was that episode which sent me back to the psychiatrist and to accept medication. The graph shows the effect of that medication. (I ought to explain that I have recalibrated from May this year: on my old system of measurement those most recent numbers would have been much tighter.)

This is my graph. It is what my GP and psychiatrist and therapist need to see. It is how my buddies keep track. It is what I need to see.

It is not anecdotal evidence. It is hard data.

The numbers speak for themselves.

Mary

A Moodscope member.

Thoughts on the above? Please feel free to post a comment below.

Moodscope members seek to support each other by sharing their experiences through this blog. Posts and comments on the blog are the personal views of Moodscope members, they are for informational purposes only and do not constitute medical advice.

Email us at support@moodscope.com to submit your own blog post!

Comments

You need to be Logged In and a Moodscope Subscriber to Comment and Read Comments