The Covid survey and your data
The ONS wants health records for another 15 years.
Last month I wrote about the GP health records data sharing scheme - and my impatience with the delays to a project where I feel the benefits outweigh the privacy concerns. So I was interested to receive an email this week from the National Statistician Professor Sir Ian Diamond wanting more access to my. data.
Just to be clear, this was not a personal email but one sent to everyone who is participating in the Office for National Statistics’ Covid-19 survey. Every month someone turns up at my door, asks me a whole series of questions - “how many times have you been to the shops this past week?” “how many people over the age of 70 have you had close contact with?” - and then gets me to swab my throat and nose for a PCR test. A few days later I get sent the result of the test (always negative so far) and a £25 voucher to spend at a wide range of retailers - which is nice.
But nicer still is that this ONS project has been a vital tool in tracking the pandemic and also, according to Sir Ian’s email, “the impact of vaccines on getting and passing on COVID-19.”
When I signed up, the email told me, I agreed that the ONS could link the data I gave them with my NHS health records and go on doing so for a year after my final visit from the researcher. This enables the statisticians to find out things like how much having COVID-19 or being vaccinated affects how often people go into hospital or visit their GP.
Now, though, Sir Ian wants an extension of that data sharing - and not a small one. The ONS wants to go on linking the survey and NHS data for 15 years after people leave the study. It also wants to look back as far as January 2016 at their health data to make sure it can take account of any health conditions they had before entering the study.
In my case, that would include the ocular melanoma diagnosed in 2005 and my Parkinson’s diagnosis from 2019. Why do they need this information? Because it will help researchers work out whether people who have had Covid-19 are more likely to get other conditions in the future.
If they don’t want this data sharing to happen, users are given an email address or a phone number they can contact to opt out. Now, this is exactly the kind of mission creep privacy campaigners warn about in data collection exercises of this kind, and I imagine quite a number of people will refuse to agree to the extension.
But I won’t be opting out. After all, the whole point of joining the survey was to give the NHS more information to plan ahead - and as it now struggles to deal with a backlog of patients with conditions other than Covid the insights the data will provide may become even more valuable. Unlike some of the government’s attempts to communicate health information during the pandemic, Sir Ian’s email is admirably clear and informative.
If you stood on your doorstep in 2020 banging pots and pans to show your appreciation for the NHS, that was a nice gesture. But handing over your data would be a lot more useful.
.
It can't be said enough, we need to encourage more people to agree to have their health data made available for research. Well done for volunteering, and promoting this.
I'm not a privacy campaigner although I have been working in this field for some years which has helped provide a perspective on the value of healthcare data to the NHS, DHSC, Local Government and external healthcare organisations. So in principle, I'm all for sharing relevant Covid-19 survey data and health information with the ONS which is looking at the spread of the virus in communities with the aim of forecasting future healthcare requirements.
However, when I see an ONS statements such as "Information that can identify you will only be held by the ONS and IQVIA for the purposes of the study", it sets alarm bells ringing. There is very little information on what personal or health information this external US data science company has access to or what permissions they have been given by ONS or indeed the NHS on what they can do with it.
The ONS survey shouldn't be conflated with the recent NHS care.data lookalike that was poorly constructed and badly publicised last year. The reasons it failed and was shelved was because the prospect of sharing data with external, unspecified organisations didn't offer sufficient safeguards i.e. the true anonymisation of records. It also offered a very short timescale within which individuals had to decide to remain in or opt out of a proposal which provided little information on what might be shared and with whom.
Covid-19 has provided an opportunity for data harvesting, the like of which we have never seen before. But as we've seen with the purchases of PPE and the government's own VIP lane, some actions have been taken in the public interest which in itself is not a bad thing but is easily corrupted. Public interest doesn't obviate the need to maintain safeguards over the privacy of individual healthcare data which has a value beyond that of the analysis of anonymised healthcare data.
How many of us scoffed at those who warned about the intrusion of Facebook into manipulating the minutiae of our personal lives to generate billions in profits at our expense?
Healthcare data is even more valuable and we make it freely available at our peril.