This is a guest blog kindly contributed by Dan Holloway in response to HEPI’s recent research on well-being. Dan sits on the Oxford University Disability Advisory Group and has been an adviser to the financial sector on debt, mental health, and financial inclusion for more than a decade. He gave the 2017 Oxford Disability Lecture on the subject of mental illness as an access issue for higher education. These are his views and not those of the University of Oxford. His website is: https://rogueinterrobang.com
I welcome any reminder that mental illness and mental health lie on a different axis from high and low levels of wellbeing. And I recognise the value of working to ensure higher levels of wellbeing across groups of people who aren’t ill, as well as those who are. To that extent I welcome both Rachel Hewitt’s policy note, and Hugo Dale-Harris’s thoughtful response to it.
I want to question two assumptions which, for all the differences between these pieces, lie behind both of them. First is the idea that while focusing on mental health will benefit a subset of the university population, focusing on wellbeing will benefit everyone. Second, I want to ask whether metrics for wellbeing might have unintended negative consequences on the mentally ill. I want to argue that those of us who are disabled or have a mental illness might want different data when it comes to choosing a university. I will offer a couple of suggestions. Most of all I want to make the very simple case that no policy be decided without involvement at every level, including a leadership one, of the disabled and the mentally ill.
The principle of universal design is very popular when it comes to making products inclusive. Build a product well, build it inclusively, the principle goes, and it will benefit everyone. I see this most in the financial services sector, but it applies across many fields. And I think this is behind the desire to measure and promoting wellbeing, when wellbeing is seen as embracing all people while mental illness only covers some. Institutions that do well at wellbeing will benefit everyone who passes through them, the argument would run. The disabled and the mentally ill might need additional steps, but they will be starting from an already-raised foundation.
The problem is this is not true. There is no universal experience in this area. As a result there can be no universal design nor measurement of universal benefit, no matter how well benchmarked. Schemes and products that benefit the wellbeing of most people can be actively harmful to the wellbeing of those with a mental illness. One example from the finance sector is the use of chatbots. Many people do not just prefer to interact with a human being, they thrive on it. Some customers, such as the elderly, depend upon it. On the other hand, customers with a mental illness will disengage more when dealing with a person than when dealing with what they know to be a machine. There is no universal design that can work around this.
This is not to deny the value of trying to improve wellbeing. But it is to say that creating incentives for those who do well at wellbeing, or who treat wellbeing as an across-the-board goal, will not necessarily also benefit the mentally ill, and may actually damage them.
The result of this is simple. In order to benefit the wellbeing of all, you must treat the wellbeing of the mentally ill differently. And, indeed, that of other groups in vulnerable positions. In other words, it is a red herring to treat wellbeing as a separate axis. The only way to benefit everyone’s wellbeing is to ignore wellbeing as a general principle and focus on the specific needs of specific groups.
That brings me to my concern about metrics in general. Everyone is familiar with the problematic results of measuring heart surgeons by their survival rates – they simply avoid taking on the most complicated patients. We can talk endlessly about benchmarking and ensuring we measure added value, but the problem remains. Metrics always have unintended consequences. And if we measure the wellbeing of students, we will create an incentive for higher education institutions to exclude those whose conditions are complex, and those whose conditions are unlikely to change. Indeed if we accept the value of metrics and algorithms, we would exclude even those who do not have a diagnosed condition but who might be part of a group more likely to have an undiagnosed condition, or to develop a condition later in life that will lead to lower wellbeing.
I am not suggesting we measure nothing. I am suggesting that universal measures may lead to unintended negative consequences for those with a mental illness.
Other measures might be possible. Every metric is a proxy for something of value to someone. As a disabled staff member, one of the things I value most is knowing that the place I work takes discrimination seriously in practice as well as principle. So, for example, the number of non-disclosure agreements undertaken alongside the percentage of grievance and harassment cases upheld would tell me an interesting story.
I would also like to read about the experiences of people like me, provided they have been told fully and without fear of consequence. In the social sciences, we rarely see a study that does not adopt a mixed methodology, combining data with narratives that allow us to understand them. I would like to see that applied here, so that potential applicants who are disabled or have a mental illness can place figures in the context of actual experiences, and can read actual experiences in the light of reliable data.
If there is going to be a set of metrics, then those likely to be impacted must be at the centre of deciding what they are. This reflects the fundamental principle “nothing about us without us.”. And it is essential in this context because disabled people, the mentally ill, the marginalised and vulnerable, have experienced these unintended consequences throughout their lives. They can identify them, and help avoid them. Many others do not have this experience. Many of those do not realise these consequences exist. Following the recommendation of the Equality and Human Rights Commission in its update report to the UN Committee for People with Disabilities , the process of deciding upon metrics, if they are to be used, must be led and governed by those directly affected.