Avoiding gibberish when assessing for dementia

language-barrier.jpgAt the inception of a new year, I’m looking back on one of my favourite pieces of work from 2017, which was commended and a joint runner up for the Max Perutz Award, detailed in my previous post:

Have you heard the saying “No ifs ands or buts”? Associated with grannies and teachers, you’d be hard pressed to find someone who hasn’t. It’s also a saying used in the ACE, a test implemented across the UK to detect dementia. An individual has to read the saying out loud with correct pronunciation. When directly translated into Urdu it loses meaning and becomes gibberish and reading out gibberish isn’t a smooth task. Even a fluent Urdu speaker might fail. But would we attribute that to dementia? Apparently, yes.

On a national level we implement a one size fits all policy regarding tests for detecting dementia that examine cognition with tasks such as the above- child’s play. But what if you’re Caucasian and I asked you to read a Chinese idiom? Or memorise words in Arabic? Or recognise an illustration of a dohl- a South Asian oblong drum? You’d probably fail on more than one task but it wouldn’t be attributed to dementia, but a mismatch in language and culture.

But when it comes to ethnic minorities we don’t account for this bias. We administer English tests, assuming everyone is fluent, and ignore the fact that they were designed for specific European cultures. When ethnic minorities can’t attempt these tests no diagnosis or treatment is given and when they manage to complete them, but fail due to this bias, they are presumed to have dementia, experiencing all the distress and depletion of time, effort and resources this diagnosis encompasses. Simultaneously, these incorrect diagnoses contribute to the already burdensome national £26 billion cost of attending to over 700,000 individuals currently diagnosed with dementia.

20180121_004142.jpgMy supervisor and I saw this particularly in British South Asians, who are more likely to slip through the cracks or receive an incorrect diagnosis in a foreign and confusing system. Therefore, they became the centrefold in my attempt to level the playing field by developing a test for indicating dementia, unbiased by language or culture. I decided to change the ACE.

I asked others who had adapted the ACE for their own countries how they did it and learnt the underlying cognitions examined by each question of the test, be it attention, memory or fluency. I devised my own process for adapting the ACE for any culture and from the vast tongues and dialects of South Asia I decided to translate it into Urdu, the UK’s fourth most spoken language. I avoided the Google translate shortcut and instead translated myself, word by word. Most importantly, I redesigned the cultural aspects of the test by including South Asians in the process.

During focus groups with South Asian elderly I showed an illustration of a kangaroo from the ACE and less than half the men and women correctly identified it. Why should they, I realised, with their roots in South Asia, where there are no kangaroos, not even in zoos. If they failed to recognise its likeness we couldn’t attribute that to dementia so we picked another animal to replace it, one they chose to be suitable. This is how these men and women narrated their culture and came up with suggestions for each question of the ACE, not changing the underlying cognitive concepts but making them understandable and acceptable.

They were eager to contribute and relayed to me in between sips of their tea that dementia was known to all of them in some form or other; a friend, a brother, a wife, all lost, never diagnosed officially or receiving help far too late. These stories unfolded as a penguin became a peacock, Harry Barnes became Haroon Butt and “No ifs ands or buts” became the better known South Asian idiom “You can’t clap with one hand”.

With our blood, sweat, tears and my homemade Urdu keyboard on which I’d drawn the curvy script the Urdu ACE was born and administered to cognitively healthy elderly South Asians. Afterwards, I interviewed them, asking about their experience taking the test. The overwhelming consensus was positive. The test was deemed “straightforward”, with “nothing vague about it” and participants felt comfortable undergoing it, claiming “everybody should be able to”. This was novel for them, a test that catered to their culture, which they could attempt with no barriers and they recognised just how important it was for their communities. One grandfatherly man handed me a celebratory candy and enthusiastically exclaimed “It will help me and people around the world for diagnosis and better treatment!

Indeed I hope it will because no one should be denied treatment for not completing a test and no one should receive a life changing diagnosis on a technicality. Is the bias eliminated? Not yet, but it’s definitely a small win while we endeavour to level the playing field for good.

*This post originally featured on the MRC’s official blog, Insight!

MRC Insight.png


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s