family, National Pupil Database, politics, pupils

Free School Meals: A political football and the need for research to referee

I wrote this post in July 2014, before the introduction of the universal infant free school meals programme (UIFSM) and before I put my interest in data to work. Here’s an updated version. My opinion why I feel it is vital that  public health and socio economic research should create an evidence base that justifies or refutes policy. 

I wondered last year whether our children’s health and the impact of UIFSM was simply a political football, which was given as a concession in the last Parliament, rushed through to get checked-off, without being properly checked out first?

How is UIFSM Entitlement Measured and What Data do we Have?

I have wondered over this year how the new policy which labels more children as entitled to free school meals may affect public health and social research.

The Free School Meal (FSM) indicator has been commonly used as a socio-economic indicator.

In fact, there is still a practical difference within the ‘free school meals’ label.

In my county, West Sussex, those who are entitled to FSM beyond infants must actively register online. Although every child in Reception, Years 1 and 2  is automatically entitled to UIFSM, parents in receipt of the state income benefits must actively register with county to have an FSM eligibility check, so that schools receive the Pupil Premium.  Strangely having to register for ‘Free School Meals’ where others need not under automatic entitlement in infants – because it’s not called as it probably should be ‘sign up for Pupil Premium’ which benefits the school budget and one hopes, the child with support or services they would not otherwise get.

Registering for a free school meal eligibility check could raise an extra grant of £1,320 per year, per child, for the child’s primary school, or £935 per child for secondary schools, to fund valuable support like extra tuition, additional teaching staff or after school activities. [source]

Researchers will need to give up the FSM indicator used as an adopted socio-economic function in age groups under 8. Over 8 (once children leave infants) only those entitled due to welfare status and actively  registered will have the FSM label. Any comparative research can only use the Pupil Premium status, but as the benefits which permit applying for it changed too, comparison will be hard. An obvious and important change to remember measuring  the effects of the policy change have had.

One year on, I’d also like to understand how research may capture the changes of children’s experience in reality.

There are challenges in this; not least getting hold of the data. Given that private providers may not all be open to provision of information, do not provide data as open data, and separately, are not subject to the Freedom of Information Act, we may not be able to find out the facts around the changes and how catering meets the needs of some of our youngest children.

If it can be hard to access information from private providers held by them, it can be even harder to do research in the public interest using information about them. In my local area Capita manages a local database and the meal providers are private companies. (No longer staff directly employed and accountable to schools as once was).

[updated Aug 30 HT Owen Boswara for the link to the Guardian article in March 2015 reporting that there are examples where this has cut the Pupil Premium uptake]

Whom does it benefit most?

Quantity or Quality and Equality?

In last year’s post I considered food quality and profit for the meal providers.

I would now be interested to see research on what changes if any there have been in the profit and costs of school meal providers since the UIFSM introduction and what benefits we see for them compared with children.

4 in 10 children are classed as living in poverty but may not meet welfare benefit criteria according to Nick Clegg, on LBC on Sept 5th 2014. That was a scandalous admission of the whole social system failure on child poverty. Hats off to the nine year-old who asked good questions last year.

The entitlement is also not applied to all primary children equally, but infants only. So within one family some children are now entitled and others are not.

I wonder if this has reshaped family evening meals for those who do not quite qualify for FSM, where now one child has already ‘had a hot meal today’ and others have not?

The whole programme of child health in school is not only unequal in application to children by age, but is not made to apply to all schools equally.

Jamie Oliver did his darnedest to educate and bring in change, showing school meals needed improvement in quality across the board. What has happened to those quality improvements he championed? Abandoned at least in free school where schools are exempt from national standards. [update: Aug 25 his recent comment].

There is clearly need when so many children are growing up in an unfairly distributed society of have and have-not, but the gap seems to be ever wider. Is Jamie right that in England eating well is a middle class concern? Is it impossible in this country to eat cheaply and eat well?

In summary, I welcome anything that will help families feed their children well. But do free school dinners necessarily mean good nutrition? The work by the Trussel Trust and others, shows what desperate measures are needed to help children who need it most and simply ‘a free school meal’ is not necessarily a ticket to good food, without rigorous application and monitoring of standards, including reviewing in schools what is offered vs what children actually eat from the offering.

Where is the analysis for people based policy that will tackle the causes of need, and assess if those needs are being met?

Evidence based understanding

It appears there were pilots and trials but we hadn’t heard much about them before September 2014. I agreed with then MP David Laws, on the closure of school kitchens, but from my own experience, the UIFSM programme lacked adequate infrastructure and education before it began.

Mr. Laws MP said,

“It is going to be one of the landmark social achievements of this coalition government – good for attainment, good for health, great for British food, and good for hard working families. Ignore the critics who want to snipe from the sidelines.”

I don’t want to be a critic from the sidelines, I’d like to be an informed citizen and a parent and know that this programme brought in good food for good health. Good for very child, but I’d like to know it brought the necessary change for the children who really needed it. [Ignoring his comment on hard working families, which indicates some sort of value judgement and out of place.]

Like these people and their FOIs, I want to ask and understand. Will this have a positive effect on the nutrition children get, which may be inadequate today?

How will we measure if UIFSM is beneficial to children who need it most?

Data used well gives insights into society that researchers should use to learn from and make policy recommendations.

The data from the meal providers and the data on UIFSM indicators as well as Pupil Premium need looked at together. That won’t be easy.

What is accessible is the data held by the DfE but that may also be “off” for true comparison because the need for active sign up is reportedly patchy.

Data on individual pupils needs used with great care due to these measurement changes in practice as well as its sensitivity. To measure that the policy is working needs careful study accounting for all the different factors that changed at the same time. The NPD has pupil premium tracked but has its uptake affected the numbers as to make it a useful comparator?

Using this administrative data  — aggregated and open data — and at other detailed levels for bona fide research is vital to understand if policies work. The use of administrative data for research has widespread public support in the public interest, as long as it is done well and not for commercial use.

To make it more usefully available, and as I posted previously, I believe the Department of Education should shape up its current practices in its capacity as the data processor and controller of the National Pupil Database to be fit for the 21st century if it is to meet public expectations of how it should be done.

Pupils and parents should be encouraged to become more aware about information used about them, in the same way that the public should be encouraged to understand how that information is being used to shape policy.

At the same time as access to state held data could be improved, we should also demand that access to information for public health and social benefit should be required from private providers. Public researchers must be prepare to stand up and defend this need, especially at a time when Freedom of Information is also under threat and should in fact expanded to cover private providers like these, not be restricted further.

Put together, this data in secure settings with transparent oversight could be invaluable in the public interest. Being seen to do things well and seeing public benefits from the data will also future-proof public trust that is vital to research. It could be better for everyone.

So how and when will we find out how the UIFSM policy change made a difference?

What did UIFSM ever do for us?

At a time when so many changes have taken place around child health, education, poverty and its measurement it is vital that public health and socio economic research creates an evidence base that justifies or refutes policy.

In some ways, neutral academic researchers play the role of referee.

There are simple practical things which UIFSM policy ignores, such as 4 year-olds starting school usually start on packed lunch only for a half term to get to grips with the basics of school, without having to manage trays and getting help to cut up food. The length of time they need for a hot meal is longer than packed lunch. How these things have affected starting school is intangible.

Other tangible concerns need more attention, many of which have been reported in drips of similar feedback such as reduced school hall and gym access affecting all primary age children (not only infants) because the space needs to be used for longer due to the increase in numbers eating hot meals.

Research to understand the availability of facilities and time spent on sport in schools since the introduction of UIFSM will be interesting to look at together with child obesity rates.

The child poverty measurements also moved this year. How will this influence our perception of poverty and policies that are designed to tackle it?

Have we got the data to analyse these policy changes? Have we got analysis of the policy changes to see if they benefit children?

As a parent and citizen, I’d like to understand who positions the goalposts in these important public policies and why.

And who is keeping count of the score?

****

image source: The Independent

refs: Helen Barnard, JRF. http://www.jrf.org.uk/blog/2015/06/cutting-child-benefit-increasing-free-childcare-where-poverty-test

#caredata, care.data, change

Building Public Trust [5]: Future solutions for health data sharing in care.data

This wraps up my series of thoughts on ‘Building Public Trust’ since the NIB Bristol meeting on July 24th.

It has looked at how to stop chasing public trust and instead the need to become organisations that can be trustworthy [part 1]. What behaviours make an organisation trustworthy [part 2]. Why fixing the Type 2 opt out is a vital first step [part 3], and why being blinded by ‘the benefits’ is not the answer [part 4], but giving balanced and fair explanations of programme purposes, commissioning and research, is beneficial to communicate.

So I want to wrap up by suggesting how communications can be improved in content and delivery. Some ideas will challenge the current approach.

Here in part five: Future solutions, I suggest why aiming to “Build Public Trust” through a new communications approach may work better for the public than the past. I’ll propose communications on care.data:

  • Review content:  what would ethical, accurate content look like
  • Strengthen relationships for delivery: don’t attempt to rebuild trust where there is now none, but strengthen the channels that are already viewed by the public to be trustworthy
  • Rethink why you communicate and the plan for when: All communications need delivered through a conversation with real listening and action based upon it. Equal priority must be given to both a communications plan for today and for the future. It must set out a mechanism for future change communications now,  before the pathfinders begin
  • Since writing this, the Leeds area CCGs have released their ‘data sharing’ comms leaflet. I have reviewed this in detail and give my opinions as a case study.

NIB workstream 4, underpins the NHS digital future,  and aims to build and sustain public trust, delivering plans for consent based information sharing and assurance of safeguards. It focuses on 4 areas: governance and oversight, project risks, consent and genomics:

“The work will begin in 2015 and is expected to include deliberative groups to discuss complex issues and engagement events, as well as use of existing organisations and ways to listen. There will also be a need to listen to professional audiences.”  [NIB work stream 4] [ref 1]

Today’s starting point in trust, trust that enables two-way communication, could hardly be worse, with professionals and public audiences. Communications are packaged in mistrust:

“Relations between the doctors’ union and Health Secretary Jeremy Hunt hit a new low following his announcement in July that he was prepared to impose seven-day working on hospital doctors in England.” [BBC news, Aug 15, 2015]

There appears to be divided opinion between politicians and civil servants.

Right now, the Department of Health seems to be sabotaging its own plans for success at every turn.

What reason can there be for denying debate in the public domain of the very plans it says are the life blood of the savings central to the NHS future?

Has the Department learned nothing from the loss of public and professional trust in 2014?

And as regards the public in engagement work, Hetan Shah, executive director of the Royal Statistical Society said in 2014, “Our research shows a “data trust deficit”. In this data rich world, companies and government have to earn citizens’ trust in how they manage and use data – and those that get it wrong will pay the price.’ [RSS Data Trust Deficit, lessons for policymakers, 2014] [2]

Where do the NIB work stream discussions want to reach by 2020?

“The emergence of genomics requires a conversation about what kind of consent is appropriate by 2020. The work stream will investigate a strand of work to be led by an ethicist.” [NIB work stream 4]

Why is genomics here in workstream 4, when datasharing for genomics is with active consent from volunteers? Why will a strand of work be led by an ethicist for this, and not other work strands? Is there a gap in how their consent is managed today or in how consent is to be handled for genomics for the future? It seems to me there is a gap in what is planned and what the public is being told here. It is high time for an overdue public debate on what future today’s population-wide data sharing programme is building. Good communication must ensure there are no surprises.

The words I underlined from the work stream 4 paper, highlight the importance of communication; to listen and to have a conversation. Despite all the engagement work of 2014 I feel that is still to happen. As one participant summed up later, “They seem hell bent on going ahead. I know they listened, but what did they hear?” [3]

care.data pathfinder practices are apparently ready to roll out communications materials: “Extraction is likely to take place between September and November depending on how fair processing testing communications was conducted” [Blackburn and Darwen HW]

So what will patient facing materials look like in content? How will they be rolled out?

Are pathfinder communications more robust than 2014 materials?

I hope the creatives will also think carefully, what is the intent of communications to be delivered.  Is it to fully and ethically inform patients about their choice whether to accept or opt out from changes in their data access, management, use and oversight? Or is the programme guidance to minimise the opt out numbers?

The participants are not signing up to a one time, single use marketing campaign, but to a lifetime of data use by third parties. Third parties who remain in role and purposes, loosely defined.

It is important when balancing this decision not to forget that data  that is available and not used wisely could fail to mitigate risk; for example in identifying pharmaceutical harms.

At the same time to collect all data for all purposes under that ‘patient safety and quality’ umbrella theme is simplistic, and lends itself in some ways, to lazy communications.

Patients must also feel free and able to make an informed decision without coercion, that includes not making opting out feel guilty.

The wording used in the past was weighted towards the organisation’s preference.  The very concept of “data sharing” is weighted positively towards the organisation. Even though in reality the default is for data to be taken by the organisation, not donated by the citizen. In other areas of life, this is recognised as an unwilling position for the citizen to be in.

At the moment I feel that the scope of purposes both today and future are not clearly defined enough in communications or plans for me personally to be able to trust them. Withholding information about how digital plans will fit into the broader NHS landscape and what data sharing will mean beyond 2020 appears rightly or wrongly,  suspicious. Department of Health, what are you thinking?

What the organisation says it will do, it must do and be seen to do, to be demonstrably trustworthy.

This workstream carries two important strands of governance and oversight which now need to be seen to happen. Implementing the statutory footing of the National Data Guardian, which has been talked about since October 2014 and ‘at the earliest opportunity’ seems to have been rather long in coming, and ‘a whole system’ that respects patient choice. What will this look like and how will it take into account the granular level of choices asked for at care.data listening events through 2014?

“By April 2016 NIB will publish, in partnership with civil society and patient leaders, a roadmap for moving to a whole-system, consent-based approach, which respects citizens’ preferences and objections about how their personal and confidential data is used, with the goal of implementing that approach by December 2020.”

‘By December 2020’ is still some time away, yet the pathfinders for care.data rolls on now regardless. The proof that will demonstrate what was said about data use actually is what happens to data, that what is communicated is trustworthy, is part of a system that can communicate this by recording and sharing consent decisions, “and can provide information on the use to which an individual’s data has been put. Over the longer term, digital solutions will be developed that automate as far as possible these processes.”

Until then what will underpin trust to show that what is communicated is done, in the short term?

Future proofing Communications must start now

Since 2013 the NHS England care.data approach appeared to want a quick data grab without long term future-proofed plans. Like the hook-up app approach to dating.

To enable the NIB 2020 plans and beyond, to safeguard research in the public interest, all communications must shape a trusted long term relationship.

To ensure public trust, communications content and delivery can only come after changes. Which is again why focusing only on communicate the benefits without discussing balance of risk does not work.  That’s what 2014 patient facing communications tried.

In 2014 there were challenges on communications that were asked but not answered, on reaching those who are digitally excluded, on reaching those for whom reading text was a challenge, and deciding who the target audience will be, considering people with delegated authority young and old, as well as those who go in and out of GP care throughout their lives, such as some military. Has that changed?

In February 2014 Health Select Committee member Sarah Wollaston, now Chair, said: “There are very serious underlying problems here that need to be addressed.”

If you change nothing, you can expect nothing to change in public and professional feeling about the programme. Communications cannot in 2015 simply revamp the layout and pacakging. There must be a change in content and in the support given in its delivery. Change means that you need to stop doing some things and start doing others.

In summary for future communications to support trust, I suggest:

1. STOP: delivering content that is biased towards what the organsation wants to achieve often with a focus on fair processing requirement, under a coercive veil of patient safety and research

START: communicating with an entirely ethical based approach reconsidering all patient data held at HSCIC and whether omission of  ‘commercial use’, balanced risks as identified in the privacy impact assessment and stating ‘your name is not included’ is right.  

2. STOP: Consider all the releases of health data held by HSCIC again and decide for each type if they are going to deliver public confidence that your organisations are trustworthy. 

START: communicate publicly which commercial companies, re-users and back office would no longer be legally eligible to receive data and why. Demonstrate organisations who received data in the past that will not in future.  

3. STOP: the Department of Health and NHS England must stop undermining trust in its own leadership, through public communications that voice opposition to medical professional bodies. Doctors are trusted much more than politicians.

START: strengthen the public-GP relationship that is already well trusted. Strengthen the GP position that will in turn support the organisational-trust-chain that you need to sustain public support. 

4. STOP: stop delaying the legislative changes needed on Data Guardian and penalties for data misuse 

START: implement them and clearly explain them in Parliament and press

5. STOP: don’t rush through short term short-cuts  to get ‘some’ data but ignore the listening from the public that asked for choice.

START: design a thorough granular consent model fit for the 21stC and beyond and explain to the public what it will offer, the buy in for bona fide research will be much greater (be prepared to define ‘research’!

6. STOP: saying that future practices have been changed and that security and uses are now more trustworthy than in the past. Don’t rush to extract data until you can prove you are trustworthy.

START: Demonstrate in future who receives data to individuals through a data use report. Who future users are in practice can only be shown through a demonstrable tool to see your word can be relied upon in practice. This will I am convinced, lower the opt out rate.

 Point 6 is apparently work-in-progress. [p58]
NIB2015

7. STOP: rolling out the current communications approach without any public position on what changes will mean they are notified before a new purpose and user in future of our data

START: design a thorough change communications model fit for the 21stC and beyond and tell the public in THIS round of communications what changes of user or purposes will trigger a notification to enable them to opt out in future BEFORE a future change i.e. in a fictional future – if the government decided that the population wide database should be further commercialised ‘for the purposes of health’, linked to the NHSBT blood donor registry and sold to genomic research companies, how would I as a donor be told, BEFORE the event?

There are still unknowns in content and future scope that mean communications are difficult. If you don’t know what you’re saying how to say it is hard. But what is certain is that there are future changes in the programme planned, and how to communicate these these with the public and professionals must be designed for now, so that what we are signed up for today, stays what we signed up for.

Delivering messages about data sharing and the broader NHS, the DH/NHS England should consider carefully their relationships and behaviours, all communication becomes relevant to trust.

Solutions cannot only be thought of in terms tools, not of what can be imposed on people, but of what can be achieved with people.

That’s people from the public and professionals and the programme working with the same understanding of the plans together, in a trusted long term relationship.

For more detail including my case study comments on the Leeds area CCGs comms leaflet, continue reading below.

Thanks for sharing in discussions of ideas in my five part post on Building public trust – a New Approach. Comments welcome.

Continue reading Building Public Trust [5]: Future solutions for health data sharing in care.data →

#caredata, care.data

Building Public Trust [4]: “Communicate the Benefits” won’t work for care.data

care.data communicating the benefits as its response to the failed communications in spring 2014, has failed to deliver public trust, here’s why:

To focus on the benefits is a shortcut for avoiding the real issues

Talking about benefits is about telling people what the organisation wants to tell them. This fails to address what the public and professionals want to know. The result is not communication, but a PR exercise.

Talking about benefits in response to the failed communications in spring 2014 and failing to address criticism since, ignores concerns that public and professionals raised at macro and micro level.  It appears disingenuous about real engagement despite saying ‘we’re listening’ and seems uncaring.

Talking about only the benefits does not provide any solution to demonstrably outweigh the potential risk of individual and public health harm through loss of trust in the confidential GP relationship, or data inaccuracy, or loss, and by ignoring these, seems unrealistic.

Talking about short term benefits and not long term solutions [to the broken opt out, long term security, long term scope change of uses and users and how those will be communicated] does not demonstrate competency or reliability.

Talking about only the benefits of commissioning, and research for the merged dataset CES, doesn’t mention all the secondary uses to which all HSCIC patient level health data are put, [those reflected in Type 2 opt out] including commercial re-use and National Back Office: “2073 releases made from the National Back Office between April 2013 and December 2013. This includes 313 releases to police forces, 1531 to the Home Office and 229 to the National Crime Agency.” [HSCIC, July2,  2014].

This use of hospital records and other secondary data by the back office, without openly telling the public, does not feel  ethical and transparent.

Another example, is the past patient communications that expressly said, ‘we do not collect name’, the intent of which would appear to be to assure patients of anonymity, without saying name is already stored at HSCIC on the Personal Demographics Service, or that name is not needed to be identifiable.

We hear a lot about transparency. But is transparent the same fully accurate, complete and honest? Honest about the intended outcomes of the programme. Honest about all the uses to which health data are put. Honest about potential future scope changes and those already planned.

Being completely truthful in communications is fundamental to future-proofing trust in the programme.

NHS England’s care.data programme through the focus on ‘the benefits’ lacks balance and appears disingenuous, disinterested,  unrealistic and lacking in reliability, competency and honesty. Through these actions it does not demonstrate the organisation is trustworthy.  This could be changed.

care.data fundamentally got it wrong with the intention to not communicate the programme at all.  It got it wrong in the tool and tone of communications in the patient leaflet.  There is a chance to get it right now, if the organisation  would only stop the focus on communicating the benefits.

I’m going to step through with a couple of examples why to-date, some communications on care.data and use of NHS data are not conducive to trust.

Communication designed to ‘future-proof’ an ongoing relationship and trust must be by design, not afterthought.

Communications need to start addressing the changes that are happening and how they make people feel and address the changes that create concern – in the public and professionals – not address the  goals that the organisation has.

Sound familiar? Communications to date have been flawed in the same way that the concept of ‘building trust’ has been flawed. It has aimed to achieve the wrong thing and with the wrong audience.

Communications in care.data needs to stop focussing on what the organisation wants from the public and professionals – the benefits it sees of getting data – and address instead firstly at a macro level, why the change is necessary and why the organisation should be trusted to bring it about.

When explaining benefits there are clearly positives to be had from using primary and secondary data in the public interest. But what benefits will be delivered in care.data that are not already on offer today?

Why if commissioning is done today with less identifiable data, can there be no alternative to the care.data level of identifiable data extraction? Why if the CPRD offers research in both primary and secondary care today, will care.data offer better research possibilities? And secondly at a micro level, must address questions individuals asked up and down the country in 2014.

What’s missing and possible to be done?

  1. aim to meet genuine ongoing communication needs not just legal data protection fair processing tick-boxes
  2. change organisational attitude that encourages people to ask what they each want to know at macro and micro level – why the programme at all, and what’s in it for me? What’s new and a benefit that differs from the status quo? This is only possible if you will answer what is asked.
  3. deliver robust explanations of the reason why the macro and micro benefits demonstrably outweigh the risk of individual potential harms
  4. demonstrate reliability, honesty, competency and you are trustworthy
  5. agree how scope changes will trigger communication to ‘future-proof’ an ongoing relationship and trust by design.

As the NIB work stream on Public Trust says, “This is not merely a technical exercise to counter negative media attention; substantial change and long-term work is needed to deliver the benefits of data use.”

If they’re serious about that long term work, then why continue to roll out pathfinder communications based on a model that doesn’t work, with an opt out that doesn’t work? Communications isn’t an afterthought to public trust. It’s key.

If you’re interested in details and my proposals for success in communications I’ve outlined in depth below:

  • Why Communicate Changes at all?
  • What is change in care.data about?
  • Is NHS England being honest about why this is hard?
  • Communicate the Benefits is not working
  • A mock case study in why ‘communicate the benefits’ will fail
  • Long term trust needs a long term communications solution
  • How a new model for NHS care.data Communication could deliver

Continue reading Building Public Trust [4]: “Communicate the Benefits” won’t work for care.data →

#caredata, care.data, NHS England

Building Public Trust in care.data datasharing [3]: three steps to begin to build trust

Let’s assume the question of public trust is as important to those behind data sharing plans in the NHS [1] as they say it is. That the success of the care.data programme today and as a result, the very future of the NHS depends upon it.

“Without the care.data programme, the health service will not have a future, said Tim Kelsey, national director for patients and information, NHS England.” [12]

And let’s assume we accept that public trust is not about the public, but about the organisation being trustworthy.[2]

The next step is to ask, how trustworthy is the programme and organisation behind care.data? And where and how do they start to build?

The table discussion on  [3] “Building Public Trust in Data Sharing”  considered  “what is the current situation?” and “why?”

What’s the current situation? On trust public opinion is measurable. The Royal Statistical Society Data Trust Deficit shows that the starting points are low with the state and government, but higher for GPs. It is therefore important that the medical profession themselves trust the programme in principle and practice. They are after all the care.data point of contact for patients.

The current status on the rollout, according to news reports, is that pathfinder  practices are preparing to rollout [4]  communications in the next few weeks. Engagement is reportedly being undertaken ‘over the summer months’. 

Understanding both public trust and the current starting point matters as the rollout is moving forwards and as leading charity and research organisation experts said: “Above all, patients, public and healthcare professionals must understand and trust the system. Building that trust is fundamental. We believe information from patient records has huge potential to save and improve lives but privacy concerns must be taken seriously. The stakes are too high to risk any further mistakes.” [The Guardian Letters, July 27, 2015]

Here’s three steps I feel could be addressed in the short term, to start to demonstrate why the public and professionals should trust  both organisation and process.

What is missing?

1. Opt out: The type 2 opt out does not work. [5]  

2 a. Professional voices called for answers and change: As mentioned in my previous summary various bodies called for change. Including the BMA whose policy [6] remains that care.data should be on a patient opt-in basis.

2b. Public voices called for answers and change: care.data’s own listening event feedback [7] concluded there was much more than ‘communicate the benefits’ that needed done. There is much missing. Such as questions on confusing SCR and care.data, legislation and concern over controlling its future change, GP concerns of their ethical stance, the Data Guardian’s statutory footing, correction of mistakes, future funding and more.
How are open questions being addressed? If at all?

3. A single clear point of ownership on data sharing and public trust communications> Is this now NIB, NHS England Patients and Information Directorate, the DH  who owns care.data now? It’s hard to ask questions if you don’t know where to go and the boards seem to have stopped any public communications. Why? The public needs clarity of organisational oversight.

What’s the Solution? 

1. Opt out: The type 2 opt out does not work. See the post graphic, the public wanted more clarity over opt out in 2014, so this needs explained clearly >>Solution: follows below from a detailed conversation with Mr. Kelsey.

2. Answers to professional opinions: The Caldicott panel,  raised 27 questions in areas of concern in their report. [8] There has not yet been any response to address them made available in the public domain by NHS England. Ditto APPG report, BMA LMC vote, and others >> Solution: publish the responses to these concerns and demonstrate what actions are being done to address them.

2b. Fill in the lack of transparency: There is no visibility of any care.data programme board meeting minutes or materials from 2015. In eight months, nothing has been published. Their 2014 proposal for transparency, appears to have come to nothing. Why?  The minutes from June-October 2014 are also missing entirely and the October-December 2014 materials published were heavily redacted. There is a care.data advisory board, which seems to have had little public visibility recently either. >> Solution: the care.data programme business case must be detailed and open to debate in the public domain by professionals and public. Scrutiny of its associated current costs and time requirements, and ongoing future financial implications at all levels should be welcomed by national, regional (CCG) and local level providers (GPs). Proactively publishing creates demonstrable reasons why both the organisation, and the plans are both trustworthy. Refusing this without clear justifications, seems counter productive, which is why I have challenged this in the public interest. [10]

3. Address public and professional confusion of ownership: Since data sharing and public trust are two key components of the care.data programme, it seems to come under the NIB umbrella, but there is a care.data programme board [9] of its own with a care.data Senior Responsible Owner and Programme Director. >> Solution: an overview of where all the different nationally driven NHS initiatives fit together and their owners would be helpful.

[Anyone got an interactive Gantt chart for all national level driven NHS initiatives?]

This would also help public and professionals see how and why different initiatives have co-dependencies. This could also be a tool to reduce the ‘them and us’ mentality. Also useful for modelling what if scenarios and reality checks on 5YFV roadmaps for example, if care.data pushes back six months, what else is delayed?

If the public can understand how things fit together it is more likely to invite questions, and an engaged public is more likely to be a supportive public. Criticism can be quashed if it’s incorrect. If it is justified criticism, then act on it.

Yes, these are hard decisions. Yes, to delay again would be awkward. If it were the right decision, would it be worse to ignore it and carry on regardless? Yes.

The most important of the three steps in detail: a conversation with Mr. Kelsey on Type 2 opt out. What’s the Solution?

We’re told “it’s complicated.” I’d say “it’s simple.” Here’s why.

At the table of about fifteen participants at the Bristol NIB event, Mr. Kelsey spoke very candidly and in detail about consent and the opt out.

On the differences between consent in direct care and other uses he first explained the assumption in direct care. Doctors and nurses are allowed to assume that you are happy to have your data shared, without asking you specifically. But he said, “beyond that boundary, for any other purpose, that is not a medical purpose in law, they have to ask you first.”

He went on to explain that what’s changed the whole dynamic of the conversation, is the fact that the current Secretary of State, decided that when your data is being shared for purposes other than your direct care, you not only have the right to be asked, but actually if you said you didn’t want it to be shared, that decision has to be respected, by your clinician.

He said: “So one of the reasons we’re in this rather complex situation now, is because if it’s for analysis, not only should you be asked, but also when you say no, it means no.”

Therefore, I asked him where the public stands with that now. Because at the moment there are ca. 700,000 people who we know said no in spring 2014.

Simply: They opted out of data used for secondary purposes, and HSCIC continues to share their data.

“Is anything more fundamentally damaging to trust, than feeling lied to?”

Mr. Kelsey told the table there is a future solution, but asked us not to tweet when. I’m not sure why, it was mid conversation and I didn’t want to interrupt:

“we haven’t yet been able to respect that preference, because technically the Information Centre doesn’t have the digital capability to actually respect it.”

He went on to say that they have hundreds of different databases and at the moment, it takes 24 hrs for a single person’s opt out to be respected across all those hundreds of databases. He explained a person manually has to enter a field on each database, to say a person’s opted out. He asked the hoped-for timing not be tweeted but explained that all those current historic objections which have been registered will be respected at a future date.

One of the other attendees expressed surprise that GP practices hadn’t been informed of that, having gathered consent choices in 2014 and suggested the dissent code could be extracted now.

The table discussion then took a different turn with other attendee questions, so I’m going to ask here what I would have asked next in response to his statement, “if it’s for analysis, not only should you be asked, but also when you say no, it means no.”

Where is the logic to proceed with pathfinder communications?

What was said has not been done and you therefore appear untrustworthy.

If there will be a future solution it will need communicated (again)?

“Trust is not about the public. Public trust is about the organisation being trustworthy.”

There needs to be demonstrable action that what the org said it would do, the org did. Respecting patient choice is not an optional extra. It is central in all current communications. It must therefore be genuine.

Knowing that what was promised was not respected, might mean millions of people choose to opt out who would not otherwise do so if the process worked when you communicate it.

Before then any public communications in Blackburn and Darwen, and Somerset, Hampshire and Leeds surely doesn’t make sense.

Either the pathfinders will test the same communications that are to be rolled out as a test for a national rollout, or they will not. Either those communications will explain the secondary uses opt out, or they will not. Either they will explain the opt out as is [type 2 not working] or as they hope it might be in future. [will be working] Not all of these can be true.

People who opt out on the basis of a broken process simply due to a technical flaw, are unlikely to ever opt back in again. If it works to starts with, they might choose to stay in.

Or will the communications roll out in pathfinders with a forward looking promise, repeating what was promised but has not yet been done? We will respect your promise (and this time we really mean it)? Would public trust survive that level of uncertainty? In my opinion, I don’t think so.

There needs to be demonstrable action in future as well, that what the org said it would do, the org did. So the use audit report and how any future changes will be communicated both seem basic principles to clarify for the current rollout as well.

So what’s missing and what’s the solution on opt out?

We’re told “it’s complicated.” I say “it’s simple.” The promised opt out must work before moving forward with anything else. If I’m wrong, then let’s get the communications materials out for broad review to see how they accommodate this and the future re-communication of  second process.

There must be a budgeted and planned future change communication process.

So how trustworthy is the programme and organisation behind care.data?

Public opinion on trust levels is measurable. The Royal Statistical Society Data Trust Deficit shows that the starting points are clear. The current position must address the opt out issue before anything else. Don’t say one thing, and do another.

To score more highly on the ‘truthworthy scale’ there must be demonstrable action, not simply more communications.

Behaviours need change and modelled in practice, to focus on people, not  tools and tech solutions, which make patients feel as if they are less important to the organisations than their desire to ‘enable data sharing’.

Actions need to demonstrate they are ethical and robust for a 21stC solution.

Policies, practical steps and behaviours all play vital roles in demonstrating that the organisations and people behind care.data are trustworthy.

These three suggestions are short term, by that I mean six months. Beyond that further steps need to be taken to be demonstrably trustworthy in the longer term and on an ongoing basis.

Right now, do I trust that the physical security of HSCIC is robust? Yes.

Do I trust that the policies in the programme would not to pass my data in the future to third party commercial pharma companies? No.
Do I believe that for enabling commissioning my fully identifiable confidential health records should be stored indefinitely with a third party? No.
Do I trust that the programme would not potentially pass my data to non-health organisations, such as police or Home Office? No.
Do I trust that the programme to tell me if they potentially change the purposes from those which they outline now ? No.

I am open to being convinced.

*****

What is missing from any communications to date and looks unlikely to be included in the current round and why that matters I address in my next post Building Public Trust [4]: Communicate the Benefits won’t work for care.data and then why a future change management model of consent needs approached now, and not after the pilot, I wrap up in [5]: Future solutions.

Continue reading Building Public Trust in care.data datasharing [3]: three steps to begin to build trust →

#caredata, care.data

Building Public Trust [2]: a detailed approach to understanding Public Trust in data sharing

Enabling public trust in data sharing is not about ‘communicating benefits’. For those interested in nitty gritty, some practical suggestions for progress in Building Public Trust in data sharing follows on from my summary after the NIB Bristol event 24/7/15.

Trust is an important if invisible currency used in the two-way transactions between an organisation and people.

So far, there have been many interactions and listening events but much of what professionals and the public called for, remains undone and public trust in the programme remains unchanged since 2014.

If you accept that it is not public trust that needs built, but the tangible trusthworthiness of an organisation, then you should also ask  what needs done by the organisation to make that demonstrable change?

What’s today’s position on Public Trust of data storage and use

Trust in the data sharing process is layered and dependent on a number of factors. Mostly [based on polls and public event feedback from 2014] “who will access my data and what will they use it for?”

I’m going to look more closely below at planned purposes: research and commissioning.

It’s also important to remember that trust is not homogeneous. Trust  is nuanced even within the single relationship between one individual and another. Trust, like consent, is stratified – you may trust the Post Office to deliver a letter or postcard, but sign up to recorded delivery for something valuable.

So for example when it comes to secondary uses data sharing, I might trust HSCIC with storing and using my health records for anonymous statistics, for analysis of immunisation and illness patterns for example. But as long as they continue to share with the Home Office, police or other loosely defined third parties [5], do I want them to have fully  identifiable data at all?

Those bodies have their own public trust issues at an all time low.

Mixing the legitimate users of health data with these Back Office punitive  uses will invite some people’s opt out who would otherwise not. Some of the very groups who need the most health and social care understanding, research and care, will be the very groups who opt out if there is a possibility of police and Home Office access by the back door. Telling traveller communities what benefits care.data will bring them is wasted effort when they see NHS health data is a police accessible register. I know. I’ve talked to some about it.

That position on data storage and use should be reconsidered if NHS England is serious that this is about health and for the benefit of individuals and communities’ well being.

What has HSCIC changed to demonstrate that  it is trustworthy?

A new physical secure setting is being built that will enable researchers to view research data but not take raw data away.

That is something they can control, and have changed, and it demonstrates they take the public seriously and we reciprocate.

That is great – demonstrable change by the organisation, inviting change in the public.

That’s practical, so what can be done on policy by NHS England/DH?

What else should be done to demonstrate policy is trustworthy?

Act on what the public and professionals asked for in 2014. [8]

Right now it feels as though in public communications that the only kind of relationship that is wanted on the part of the leadership is a one night stand.

It’s all about what the programme wants. Minimise the objections, get the data, and sneak out. Even when its leaders talk about some sort of ongoing consent model, the focus is still about ‘how to enable sharing data.’

This focus is the wrong one. If you want to encourage people to share they need to know why, what’s in it for them, and why do you want it? What collecting the data is about is still important to explain and specifically, each time the scope changes if you are doing it fairly.

Remember. Data-sharing is not vital to future-proof the NHS. Using knowledge wisely is. 

What is the policy for the future of primary care research?

The CPRD already enables primary care GP data to be linked with secondary data for research. In fact it already links more items from GP held data than current are.data plans to extract. So what benefit will care.data offer to research that is not already available today?

Simply having ever more data, stored in more places will not make us wiser. Before it’s collected repeatedly, it is right to question why.

What do we have collected already? How is it used? Where are the gaps in what we want to achieve through the knowledge we could gain. It’s NOT simply about filling in what gaps exist in what data we could gather. Understand the purposes and what will be gained to see if it’s worth the efforts. Prioritise. Collect it all, is not a solution.

I had thought that the types of data to be collected in care.data were clear, and how it differs from direct care was clear. But the Bristol NIB meeting demonstrated a wide range of understanding in NHS and CCG staff, Local Authority staff, IT staff, IG professionals, data providers and other third parties.  Data for secondary purposes are not to be conflated with direct care.

But that’s not what care.data sharing is about. So where to start with public trust, asked the NIB Bristol #health2020 meeting?

Do you ignore the starting point or tailor your approach to it?

“The NHS is at a crossroads and needs to change and improve as it moves forward. That was the message from NHS England’s Chief Executive Simon Stevens as a Five Year Forward View for the NHS was launched.”  [1] [NHS England, Oct 2014]

As the public is told over and over again that change is vital to the health of a sustainable NHS, a parallel public debate rages, whether the policy-making organisations behind the NHS – the commissioning body NHS England, the Department of Health and Cabinet Office – are serious about the survival of universal health and care provision, and about supporting its clinicians.

It is against this backdrop, and under the premise that obtaining patient data for centralised secondary uses is do or die for the NHS, that the NIB #health2020 has set out [2] work stream 4: “Build and sustain public trust: Deliver roadmap to consent based information sharing and assurance of safeguards”

“Without the care.data programme, the health service will not have a future, said Tim Kelsey, national director for patients and information, NHS England.” [3]

 

Polls say [A] nearly all institutions suffer from a ‘trust in data deficit’. Trust in them to use data appropriately, is lower than trust in the organisation generally.

Public trust in what the Prime Minister says on health is low.

Trust in the Secretary of State for Health is possibly at an all time low, with: “a bitter divide, a growing rift between the Secretary of State for Health and the medical profession.” [New Statesman, July 2015]

This matters. care.data needs the support of professionals and public.

ADRN research showed multiple contributing factors: “Participants were also worried about personal data being leaked, lost, shared or sold by government departments to third parties, particularly commercial companies. Low trust in government more generally seemed to be driving these views.” [Dialogue on data]

It was interesting to see all the same issues as reflected by the public in care.data listening events, asked from the opposite perspective from data users.

But it was frustrating to sit ay the Bristol NIB #health2020 event and discuss questions around the same issues on data sharing already discussed at care.data events through the last 18 months.

Nothing substantial has changed other then HSCIC’s physical security for data storage.

It is frustrating knowing that these change and communications issues will keep coming back again and again if not addressed.

Personally, I’m starting to lose trust there is any real intention for change, if senior leadership is unwilling to address this properly and change themselves.

To see a change in Public Trust do what the public asked to see change: On Choice

At every care.data meeting I attended in 2014, people asked for choice.

They asked for boundaries between the purposes of data uses, real choice.

Willingness for their information to be used by academic researchers in the public interest does not equate to being willing for it to be used by a pharmaceutical company for their own market research and profit.

The public understand these separations well. To say they do not, underestimates people and does not reflect public feeling. Anyone attending 2014 care.data events, has heard many people discuss this. They want a granular consent model.

This would offer a red line between how data are used for what purposes.

Of the data-sharing organisations today some are trusted and others are not. Offering a granular consent approach would offer a choice of a red line between who gets access to data.

This choice of selective use, would encourage fewer people to opt out from all purposes, allowing more data to be available for research for example.

To see a change in Public Trust do what the public asked to see: Explain your purposes more robustly

Primarily this data is to be used and kept indefinitely for commissioning purposes. Research wasn’t included as purposes for care.data gathering  in the planned specifications for well over a year. [After research outcry]

Yet specific to commissioning, the Caldicott recommendations [3] were very clear; commissioning purposes were insufficient and illegal grounds for sharing fully identifiable data which was opposed by NHS England’s Commissioning Board:

“The NHS Commissioning Board suggested that the use of personal confidential data for commissioning purposes would be legitimate because it would form part of a ‘consent deal’ between the NHS and service users. The Review Panel does not support such a proposition. There is no evidence that the public is more likely to trust commissioners to handle personal confidential data than other groups of professionals who have learned how to work within the existing law.”

NHS England seems unwilling to change this position, despite the professionals bodies and the public’s opposition to sharing fully identifiable data for commissioning purposes [care.data listening events 2014]. Is it any wonder that they keep hitting the same barrier? More people don’t want that to happen than you do. Something’s gotta give.

Ref the GPES Customer Requirements specification from March 2013 v2.1 which states on page 11: “…for commissioning purposes, it is important to understand activity undertaken (or not undertaken) in all care settings. The “delta load” approach (by which only new events are uploaded) requires such data to be retained, to enable subsequent linkage.”

The public has asked for red lines to differentiate between the purposes of data uses. NHS England and the Department of Health policy seems unwilling to do so.  Why?

To see a change in Public Trust do what the public asked to see: Red lines on policy of commercial use – and its impact on opt out

The public has asked for red lines outlawing commercial exploitation of their data. Though it was said it was changed, in practice it is hard to see. Department of Health policy seems unwilling to be clear, because the Care Act 2012 purposes remained loose.  Why?

As second best, the public has asked for choice not to have their data used at all for secondary purposes and were offered an opt out.

NHS England leaflet and the Department of Health, Secretary of State publicly promised this but has been unable to implement it and to date has made no public announcement on when it will be respected.  Why?

Trust does not exist in a vacuum.  What you say and what you actually do, matter. Policy and practice are co-dependent. Public trust depends on your organisations being trustworthy.

Creating public trust is not the government, the DH or NIB’s task ahead. They must instead focus on improving their own competency, honesty and reliability and through those, they will demonstrate that they can be trusted.

That the secondary purposes opt out has not been respected does not demonstrate those qualities.

“Trust is not about the public. Public trust is about the organisation being trustworthy.”

How will they do that?

Let the DH/NHS England and organisations in policy and practice address what they themselves will stop and start doing to bring about change in their own actions and behaviours.

Communications change request: Start by addressing the current position NOT what the change will bring. You must move people along the curve , not dump them with a fait accomplice and wonder why the reaction is so dire.

changecurve

Vital for this is the current opt out; what was promised and what was done.

The secondary uses opt out must be implemented with urgency.

To see a change in Public Trust you need to take action. the Programme needs to do what the public asked to see change: on granular consent, on commercial use and defined purposes.

And to gather suggested actions, start asking the right questions.

Not ‘how do we rebuild public trust?’ but “how can we demonstrate that we are trustworthy to the public?”

1. How can a [data-sharing] org demonstrate it is trustworthy?
2. Identify: why people feel confident their trust is well placed?
3. Why do clinical professionals feel confident in any org?
4. What would harm the organisational-trust-chain in future?
5. How will the org-trust-chain be positively maintained in future?
6. What opportunities will be missed if that does not happen?
(identify value)

Yes the concepts are close,  but how it is worded defines what is done.

These apparent small differences make all the difference in how people provide you ideas, how you harness them into real change and improvement.

Only then can you start understanding why “communicating the benefits” has not worked and how it should affect future communications  materials.

From this you will find it much easier to target actual tasks, and short and long term do-able solutions than an open discussion will deliver. Doing should  include thinking/attitudes as well as actions.

This will lead to communications messages that are concrete not wooly. More about that in the next posts.

####

To follow, for those interested in nitty gritty, some practical suggestions for progress in Building Public Trust in data sharing:

Part one: A seven step top line summary – What I’d like to see change addressing public trust in health data sharing for secondary purposes.

This is Part two: a New Approach is needed to understanding Public Trust For those interested in a detailed approach on Trust. What Practical and Policy steps influence trust. On Reserach and Commissioning. Trust is not homogeneous. Trust  is nuanced even within the single relationship between one individual and another. It doesn’t exist in a vacuum.

Part three: Know where you’re starting from. What behaviours influence trust and how can we begin to see them demonstrated. Mr.Kelsey discusses  consent and opt out. Fixing what has already been communicated is vital before new communications get rolled out. Vital to tailor the content of public communications, for public trust and credibility the programme must be clear what is missing and what needs filled in. #Health2020 Bristol NIB meeting.

Part four: “Communicate the Benefits” won’t work – How Communications influence trust. For those interested in more in-depth reasons, I outline in part two why the communications approach is not working, why the focus on ‘benefits’ is wrong, and fixes.

Part five: Future solutions – why a new approach may work better for future trust – not to attempt to rebuild trust where there is now none, but strengthen what is already trusted and fix today’s flawed behaviours; honesty and reliability, that  are vital to future proofing public trust.

 

####

References:

[1] NHS England October 2014 http://www.england.nhs.uk/2014/10/23/nhs-leaders-vision/

[2] Workstream 4: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/442829/Work_Stream_4.pdf

[3] Caldicott Review 2: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/192572/2900774_InfoGovernance_accv2.pdf

[4] Missing Programme Board documents: 2015 and June-October 2014

[5] HSCIC Data release register

[6] Telegraph article on Type 2 opt out http://www.telegraph.co.uk/news/health/news/11655777/Nearly-1million-patients-could-be-having-confidential-data-shared-against-their-wishes.html

[7] Why Wanting a Better Care.Data is not Luddite: http://davidg-flatout.blogspot.co.uk/2014/04/why-wanting-better-caredata-is-not.html

[8] Talking to the public about using their data is crucial- David Walker, StatsLife http://www.statslife.org.uk/opinion/1316-talking-to-the-public-about-using-their-data-is-crucial

[9] Dame Fiona Caldicott appointed in new role as National Data Guardian

[10] Without care.data health service has no future says director http://www.computerweekly.com/news/2240216402/Without-Caredata-we-wont-have-a-health-service-for-much-longer-says-NHS

[11] Coin Street, care.data advisory meeting, September 6th 2014: https://storify.com/ruth_beattie/care-data-advisory-group-open-meeting-6th-septembe

[12] Public questions unanswered: https://jenpersson.com/pathfinder/

#caredata, care.data, change, change management, confidentiality, NHS England

Building Public Trust in care.data sharing [1]: Seven step summary to a new approach

Here’s my opinion after taking part in the NIB #health2020 Bristol event 24/7/2015 and presentation of plans at the June King’s Fund hosted event. Data sharing includes plans for extraction and uses of primary care data by third parties, charging ahead under the care.data banner.

Wearing my hat from a previous role in change management and communications, I share my thoughts in the hope the current approach can adapt and benefit from outside perspectives.

The aim of “Rebuilding and sustaining Public trust” [1] needs refocused to treat the cause, not only the symptoms of the damage done in 2014.  Here’s why:

A Seven Step Top Line Summary

1. Abstract ‘public trust’ is not vital to the future of data sharing. Being demonstrably worthy of public trust is.

2. Data-sharing is not vital to future-proof the NHS. Using knowledge wisely is.

3. A timed target to ‘get the public’s data’, is not what is needed. Having a stable, long term future-proofed and governable model is.

4. Tech solutions do not create trust. Enable the positive human response to what the org wants from people, enabling their confident ‘yes to data-sharing.’ [It might be supported by technology-based tools.]

5. Communications that tell the public ‘we know best, trust us’ fail.  While professional bodies [BMA [2], GPES advisory group, APPG report calling for a public benefits plan, ICO, and expert advice such as Caldicott] are ignored or remain to be acted upon, it remains challenging for the public to see how the programme’s needs, motives and methods are trustworthy. “The [Caldicott 2] Review Panel found that commissioners do not need dispensation from confidentiality, human rights & data protection law.” [3] Something’s gotta give. What will it be?

6. care.data consistency. Relationships must be reliable and have integrity.
“Trust us – see the benefits” [But we won’t share the business cost/benefit plan.]
“Trust us – we’re transparent” [But there is nothing published in 2015 at all from the programme board minutes] [4]
“Trust us – we’ll only use your data wisely, with the patient in control” [Ignore that we didn’t before [5] and that we still share your data for secondary uses even if you opted out [6] and no, we can’t tell you when it will be fixed…]

7. Voices do not exist in a vacuum. Being trustworthy on care.data  does not stand alone but is part of the NHS ‘big picture’.
Department of Health to GPs: “Trust us about data sharing.’  [And ignore that we haven’t respected many of  your judgement or opinions.]
NHS England to GPs: “Trust us about data sharing.’  [And ignore our lack of general GP support: MPIG withdrawal, misrepresentation in CQC reports] NHS England and Department of Health to professionals and public: “The NHS is safe in our hands.’ Everyone: “We see no evidence that plans for cost savings, 7 day working, closures and the 5YFV integration will bring the promised benefits. Let us ‘see the holes’, so that we can trust you based on evidence.”

See the differences?

Target the cause not Symptom:

The focus in the first half, the language used by NHS England/NIB/ DH, sets out their expectations of the public. “You must trust us and how you give us your data.”

The focus should instead to be on the second half, a shift to the organisation, the NHS England/NIB/ DH, and set out expectations from the public point-of-view. ” Enable the public to trust the organisation. Enable individual citizens to trust what is said by individual leaders. This will enable citizens to be consensual sharers in the activity your organisation imposes – the demand for care.data through a statutory gateway, obliging GPs to disclose patient data.

The fact that trust is broken, and specifically to data-sharing that there is the deficit [A] between how much the public trusts the organisation and how the organisation handles data, is not the fault of the public, or “1.4 M NHS staff”, or the media, or patient groups’ pressure. It’s based on proven experience.

It’s based on how organisations have handled data in the past. [5] Specifically on the decisions made by DH, and the Information Centre and leaders in between. Those who chose to sell patient data without asking the public.

The fact that trust is broken is based on how leadership individuals in those organisations have responded to that. Often taking no responsibility for loss.

No matter how often we hear “commissioners will get a better joined up picture of care needs and benefit you”, it does not compensate for past failings.

Only demonstrable actions to show why it will not happen in future can start that healing process.

Target the timing to the solution, not a shipping deadline

“Building trust to enable data sharing” aims at quick fixes, when what is needed is a healing process and ongoing relationship maintenance.

Timing has to be tailored to what needs done, not an ‘artificial deadline’. Despite that being said it doesn’t seem to match reality.

Addressing the Symptoms and not the Cause, will not find a Cure

What needs done?

Lack of public trust, the data trust deficit [A] are symptoms in the public to be understood. But it is the causes in the organisations that must be treated.

So far many NHS England staff I have met in relation to care.data, appear to have a “them and us” mentality. It’s almost tangibly wrapped up in the language used at these meetings or in defensive derision of public concerns: “Tin foil hat wearers”, “Luddites” [7] and my personal favourite, ‘Consent fetishists.’ [8] It’s counter productive and seems borne from either a lack of understanding, or frustration.

The NIB/DH/NHS England/ P&I Directorate must accept they cannot force any consensual change in an emotion-based belief based on past experiences, held by the public.

Those people each have different starting points of knowledge and beliefs.  As one attendee said, “There is no single patient replicated 60 million times.”

The NIB/DH/NHS England/ P&I Directorate can only change what they themselves can control. They have to model and be seen to model change that is trustworthy.

How can an organisation demonstrate it is trustworthy?

This means shifting the focus of the responsibility for change from public and professionals, to leadership organisation.

There is a start in this work stream, but there is little new that is concrete.

The National Data Guardian (NDG) role has been going to be put on a legal footing “at the earliest opportunity” since November 2014. [9] Nine months.

Updated information governance guidance is on the way.

Then there’s two really strong new items that would underpin public trust, to be planned in a ‘roadmap’: the first a system that can record and share consent decisions and the second, to provide information on the use to which an individual’s data has been put.

How and when those two keystones to public trust will be actually offered appear unknown. They will  encourage public trust by enabling choice and control of our data. So I would ask, if we’re not there yet on the roadmap, how can consent options be explained to the public in care.data communications, if there is as yet no mechanism to record and effect them? More on that later.

Secondly, when will a usage report be available? That will be the proof to demonstrate that what was offered, was honoured. It is one of the few tools the organisation(s) can offer to demonstrate they are trustworthy: you said, we did. So again, why jeopardise public trust by rolling out data extractions into the existing, less trustworthy environment?

How well this is done will determine whether it can realise its hoped for benefits. How the driving leadership influences that outcome, will be about the organisational approach to opt out, communicating care.data content decisions, the way and the channels in which they are communicated, accepting what has not worked to date and planning long-term approaches to communicating change before you start the pathfinders. [Detailed steps on this follows.]

Considering the programme’s importance we have been told, it’s vital to get right. [10]

i believe changing the approach from explaining benefits and focus on public trust, to demonstrating why the public should trust demonstrable changes made, will make all the difference.

So before rolling out next data sharing steps think hard what the possible benefits and risks will be, versus waiting for a better environment to do it in.

Conclusion: Trust is not about the public. Public trust is about the organisation being trustworthy. Over to you, orgs.

####

To follow, for those interested in nitty gritty, some practical suggestions for progress in Building Public Trust in data sharing:

This is Part one: A seven step top line summary – What I’d like to see change addressing public trust in health data sharing for secondary purposes.

Part two: a New Approach is needed to understanding Public Trust For those interested in a detailed approach on Trust. What Practical and Policy steps influence trust. On Research and Commissioning. Trust is not homogeneous. Trust  is nuanced even within the single relationship between one individual and another. It doesn’t exist in a vacuum.

Part three: Know where you’re starting from. What behaviours influence trust. Fixing what has already been communicated is vital before new communications get rolled out. Vital to content of your communications and vital for public trust and credibility.

Part four: Communicate the Benefits won’t work – How Communications influence trust. For those interested in more in-depth reasons, I outline in part two why the communications approach is not working, why the focus on ‘benefits’ is wrong, and fixes.

Part five: Future solutions – why a new approach may work better for future trust – not to attempt to rebuild trust where there is now none, but strengthen what is already trusted and fix today’s flawed behaviours; honesty and reliability, that  are vital to future proofing trust

####

Background References:

I’m passionate about people using technology to make their jobs and lives better, simpler, and about living well. So much so, that this became over 5000 words. To solve that, I’ve assumed a baseline knowledge and I will follow up with separate posts on why a new approach is needed to understanding “Public Trust”, to “Communicating the benefits” and “Being trustworthy and other future solutions”.

If this is all new, welcome, and I suggest you look over some of the past 18 months posts that include  public voice captured from eight care.data  events in 2014. care.data is about data sharing for secondary purposes not direct care.

[1] NHS England October 2014 http://www.england.nhs.uk/2014/10/23/nhs-leaders-vision/

[2] BMA LMC Vote 2014 http://bma.org.uk/news-views-analysis/news/2014/june/patients-medical-data-sacrosanct-declares–bma

[3] Caldicott Review 2: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/192572/2900774_InfoGovernance_accv2.pdf

[4] Missing Programme Board documents: 2015 and June-October 2014

[5] HSCIC Data release register

[6] Telegraph article on Type 2 opt out http://www.telegraph.co.uk/news/health/news/11655777/Nearly-1million-patients-could-be-having-confidential-data-shared-against-their-wishes.html

[7] Why Wanting a Better Care.Data is not Luddite: http://davidg-flatout.blogspot.co.uk/2014/04/why-wanting-better-caredata-is-not.html

[8] Talking to the public about using their data is crucial- David Walker, StatsLife http://www.statslife.org.uk/opinion/1316-talking-to-the-public-about-using-their-data-is-crucial

[9] Dame Fiona Caldicott appointed in new role as National Data Guardian

[10] Without care.data health service has no future says director http://www.computerweekly.com/news/2240216402/Without-Caredata-we-wont-have-a-health-service-for-much-longer-says-NHS

Polls of public feeling:

[A] Royal Statistical Society Data Trust Deficit http://www.statslife.org.uk/news/1672-new-rss-research-finds-data-trust-deficit-with-lessons-for-policymakers

(B] Dialogue on data – work carried out through the ADRN

 

 

#caredata, care.data, change, trust

care.data : the economic value of data versus the public interest?

 This is a repost of my opinion piece published in StatsLife in June 2015.

The majority of the public supports the concept of using data for public benefit.[1] But the measurable damage done in 2014 to the public’s trust in data sharing [2] and reasons for it, are an ongoing threat to its achievement.

Rebuilding trust and the public legitimacy of government data gathering could be a task for Sisyphus, given the media atmosphere clouded by the smoke and mirrors of state surveillance. As Mark Taylor, chair of the NHS’s Confidentiality Advisory Group wrote when he considered the tribulations of care.data [3] ‘…we need a much better developed understanding of ‘the public interest’ than is currently offered by law.’

So what can we do to improve this as pilot sites move forward and for other research? Can we consistently quantify the value of the public good and account for intangible concerns and risks alongside demonstrable benefits? Do we have a common understanding of how the public feels what is in its own best interests?

And how are shifting public and professional expectations to be reflected in the continued approach to accessing citizens’ data, with the social legitimacy upon which research depends?

Listening and lessons learned

Presented as an interval to engage the public and professionals, the 18 month long pause in care.data involved a number of ‘listening’ events. I attended several of these to hear what people were saying about the use of personal health data. The three biggest areas of concern raised frequently [4] were:

  • Commercial companies’ use and re-use of data
  • Lack of transparency and control over who was accessing data for what secondary purposes, and
  • Potential resulting harms: from data inaccuracy, loss of trust and confidentiality, and fear of discrimination.

It’s not the use of data per se that the majority of the public raises objection to. Indeed many people would object if health data were not used for research in the public interest. Objections were more about the approach to this in the past and in the future.

There is a common understanding of what bona fide research is, how it serves the public interest, and polls confirm a widespread acceptance of ‘reasonable’ research use of data. The HSCIC audit under Sir Nick Partridge [5] acknowledged that some past users or raw data sharing had not always met public expectations of what was ‘reasonable’. The new secure facility should provide a safe setting for managing this better, but open questions remain on governance and transparency.

As one question from a listening event succinctly put it [6]:

‘Are we saying there will be only clinical use of the data – no marketing, no insurance, no profit making? This is our data.’

Using the information gleaned from data was often seen as exploitation when used in segmenting the insurance markets, consumer market research or individual targeting. There is also concern, even outright hostility, to raw health data being directly sold, re-used or exchanged as a commodity – regardless whether this is packaged as ‘for profit’ or ‘covering administrative costs’.

Add to that, the inability to consent to, control or find out who uses individual level data and for what purpose, or to delete mistakes, and there is a widespread sense of disempowerment and loss of trust.

Quantifying the public perception of care.data’s value

While the pause was to explain the benefits of the care.data extraction, it actually seemed clear at meetings that people already understood the potential benefits. There is clear public benefit to be gained for example, from using data as a knowledge base, often by linking with other data to broaden scientific and social insights, generating public good.

What people were asking, was what new knowledge would be gained that isn’t gathered from non-identifiable data already? Perhaps more tangible, yet less discussed at care.data events, is the economic benefits for commissioning use by using data as business intelligence to inform decisions in financial planning and cost cutting.

There might be measurable economic public good from data, from outside interests who will make a profit by using data to create analytic tools. Some may even sell information back into the NHS as business insights.

Care.data is also to be an ‘accelerator’ for other projects [7]. But it is hard to find publicly available evidence to a) support the economic arguments for using primary care data in any future projects, and b) be able to compare them with the broader current and future needs of the NHS.

A useful analysis could find that potential personal benefits and the public good overlap, if the care.data business case were to be made available by NHS England in the public domain. In a time when the NHS budget is rarely out of the media it seems a no-brainer that this should be made open.

Feedback consistently shows that making money from data raises more concern over its uses. Who all future users might be remains open as the Care Act 2014 clause is broadly defined. Jamie Reed MP said in the debate [8]: ‘the new clause provides for entirely elastic definitions that, in practice, will have a limitless application.’

Unexpected uses and users of public data has created many of its historical problems. But has the potential future cost of ‘limitless’ applications been considered in the long term public interest? And what of the confidentiality costs [9]? The NHS’s own Privacy Impact Assessment on care.data says [10]:

‘The extraction of personal confidential data from providers without consent carries the risk that patients may lose trust in the confidential nature of the health service.‘

Who has quantified the cost of that loss of confidence and have public and professional opinions been accounted for in any cost/benefit calculations? All these tangible and intangible factors should be measured in calculating its value in the public interest and ask, ‘what does the public want?’ It is after all, our data and our NHS.

Understanding shifting public expectations

‘The importance of building and maintaining trust and confidence among all stakeholder groups concerned – including researchers, institutions, ethical review boards and research participants – as a basis for effective data sharing cannot be overstated.’ – David Carr, policy adviser at the Wellcome Trust [11]

To rebuild trust in data sharing, individuals need the imbalance of power corrected, so they can control ‘their data’. The public was mostly unaware health records were being used for secondary purposes by third parties, before care.data. In February 2014, the secretary of state stepped in to confirm an opt-out will be offered, as promised by the prime minister in his 2010 ‘every patient a willing research patient’ speech.

So leaving aside the arguments for and against opt-in versus opt-out (and that for now it is not technically possible to apply the 700,000 opt-outs already made) the trouble is, it’s all or nothing. By not offering any differentiation between purposes, the public may feel forced to opt-out of secondary data sharing, denying all access to all their data even if they want to permit some uses and not others.

Defining and differentiating secondary uses and types of ‘research purposes’ could be key to rebuilding trust. The HSCIC can disseminate information ‘for the purposes of the provision of health care or adult social care, or the promotion of health’. This does not exclude commercial use. Cutting away commercial purposes which appear exploitative from purposes in the public interest could benefit the government, commerce and science if, as a result, more people would be willing to share their data.

This choice is what the public has asked for at care.data events, other research events [12] and in polls, but to date has yet to see any move towards. I feel strongly that the government cannot continue to ignore public opinion and assume its subjects are creators of data, willing to be exploited, without expecting further backlash. Should a citizen’s privacy become a commodity to put a price tag on if it is a basic human right?

One way to protect that right is to require an active opt-in to sharing. With ongoing renegotiation of public rights and data privacy at EU level, consent is no longer just a question best left ignored in the pandora’s box of ethics, as it has been for the last 25 years in hospital data secondary use. [13]

The public has a growing awareness, differing expectations, and different degrees of trust around data use by different users. Policy makers ignoring these expectations, risk continuing to build on a shaky foundation and jeopardise the future data sharing infrastructure. Profiting at the expense of public feeling and ethical good practice is an unsustainable status quo.

Investing in the public interest for future growth

The care.data pause has revealed differences between the thinking of government, the drivers of policy, the research community, ethics panels and the citizens of the country. This is not only about what value we place on our own data, but how we value it as a public good.

Projects that ignore the public voice, that ‘listen’ but do not act, risk their own success and by implication that of others. And with it they risk the public good they should create. A state which allows profit for private companies to harm the perception of good research practice sacrifices the long term public interest for short term gain. I go back to the words of Mark Taylor [3]:

‘The commitment must be an ongoing one to continue to consult with people, to continue to work to optimally protect both privacy and the public interest in the uses of health data. We need to use data but we need to use it in ways that people have reason to accept. Use ‘in the public interest’ must respect individual privacy. The current law of data protection, with its opposed concepts of ‘privacy’ and ‘public interest’, does not do enough to recognise the dependencies or promote the synergies between these concepts.’ 

The economic value of data, personal rights and the public interest are not opposed to one another, but have synergies and a co-dependency. The public voice from care.data listening could positively help shape a developing consensual model of data sharing if the broader lessons learned are built upon in an ongoing public dialogue. As Mark Taylor also said, ‘we need to do this better.’

*******

[1] according to various polls and opinions gathered from my own discussions and attendance at care.data events in 2014 [ refs: 2, 4. 6. 12]

[2] The data trust deficit, work by the Royal Statistical Society in 2014

[3] M Taylor, “Information Governance as a Force for Good? Lessons to be Learnt from Care.data”, (2014) 11:1 SCRIPTed 1 http://script-ed.org/?p=1377

[4] Communications and Change – blogpost https://jenpersson.com/care-data-communications-change/

[5] HSCIC audit under Sir Nick Partridge https://www.gov.uk/government/publications/review-of-data-releases-made-by-the-nhs-information-centre

[6] Listening events, NHS Open Day blogpost https://jenpersson.com/care-data-communications-core-concepts-part-two/

[7] Accelerator for projects mentioned include the 100K Genomics programme https://www.youtube.com/watch?v=s8HCbXsC4z8

[8] Hansard http://www.publications.parliament.uk/pa/cm201314/cmhansrd/cm140311/debtext/140311-0002.htm

[9] Confidentiality Costs; StatsLife http://www.statslife.org.uk/opinion/1723-confidentiality-costs

[10] care.data privacy impact assessment Jan 2014 [newer version has not been publicly released] http://www.england.nhs.uk/wp-content/uploads/2014/01/pia-care-data.pdf

[11] Wellcome Trust http://blog.wellcome.ac.uk/2015/04/08/sharing-research-data-to-improve-public-health/

[12]  Dialogue on Data – Exploring the public’s views on using linked administrative data for research purposes: https://www.ipsos-mori.com/researchpublications/publications/1652/Dialogue-on-Data.aspx

[13] HSCIC Lessons Learned http://www.hscic.gov.uk/article/4780/HSCIC-learns-lessons-of-the-past-with-immediate-programme-for-change

The views expressed in this article originally published in the Opinion section of StatsLife are solely mine, the original author. These views and opinions do not necessarily represent those of The Royal Statistical Society.

National Pupil Database, privacy, pupils, scope creep

The National Pupil Database end of year report: an F in Fair Processing

2 Comments

National Pupil Database? What National Pupil Database? Why am I on it?

At the start of the school year last September 2014, I got the usual A4 pieces of paper. Each of my children’s personal details, our home address and contact details, tick boxes for method of transport each used to get to school, types of school meal eaten all listed, and a privacy statement at the bottom:

“Data Protection Act 1988: The school is registered under the Data Protection Act for holding personal data. The school has a duty to protect this information and to keep it up to date. The school is required to share some of the data with the Local Authority and with the DfE.”

There was no mention of the DfE sharing it onwards with anyone else. But they do, through the National Pupil Database [NPD] and  it is enormous [1].  It’s a database which holds personal information of every child who has ever been in state education since 2002, some data since 1996. [That includes me as both a student AND a parent.]

“Never heard of it?”

Well neither have I from my school, which is what I pointed out to the DfE in September 2014.

School heads, governors, and every parent I have spoken with in my area and beyond, are totally unaware of the National Pupil database. All are surprised. Some are horrified at the extent of data sharing at such an identifiable and sensitive level, without school and parental knowledge.[2]

Here’s a list what it holds. Fully identifiable data at unique, individual level. Tiered from 1-4, where 1 is the most sensitive. A full list of what data is available in each of the tiers and standard extracts can be found in the ‘NPD data tables’.

K5

I’d like to think it has not been deliberately hidden from schools and parents. I hope it has simply been careless about its communications.

Imagine that the data once gathered only for administration since 1996, was then decided about at central level and they forgot to tell the people whom they should have been asking. The data controllers and subjects the data were from – the schools, parents/guardians and pupils – were forgotten. That could happen when you see data as a commodity and not as people’ s personal histories.

The UK appears to have gathered admin data for years until the coalition decided it was an asset it could further exploit. The DfE may have told others in 2002 and in 2012 when it shaped policy on how the NPD would be used, but it forgot to tell the children whose information it is and used them without asking. In my book, that’s an abuse of power and misuse of data.

It seems to me that current data policies in practice across all areas of government have simply drifted at national level towards ever greater access by commercial users.

And although that stinks, it has perhaps arisen from lack of public transparency and appropriate oversight, rather than some nefarious intent.

Knowingly failing to inform schools, pupils and guardians how the most basic of our personal data are used is outdated and out of touch with public feeling. Not to mention, that it fails fair processing under Data Protection law.

Subject Access Request – User experience gets an ‘F’ for failing

The submission of the school census, including a set of named pupil records, is a statutory requirement on schools.

This means that children and parents data, regardless of how well or poorly informed they may be, are extracted for administrative purposes, and are used in addition to those we would expect, for various secondary reasons.

Unless the Department for Education makes schools aware of the National Pupil Database use and users, the Department fails to provide an adequate process to enable schools to meet their local data protection requirements. If schools don’t know, they can’t process data properly.

So I wrote to the Department for Education (DfE) in September 2014, including the privacy notice used in schools like ours, showing it fails to inform parents how our children’s personal data and data about us (as related parent/guardians) are stored and onwardly used by the National Pupil Database (NPD). And I asked three questions:

1. I would like to know what information is the minimum you require for an individual child from primary schools in England?

2. Is there an opt out to prevent this sharing and if so, under what process can parents register this?

3. Is there a mechanism for parents to restrict the uses of the data (i.e. opt out our family data) with third parties who get data from the National Pupil Database?

I got back some general information, but no answer to my three questions.

What data do you hold and share with third parties about my children?

In April 2015 I decided to find out exactly what data they held, so I made a subject access request [SAR], expecting to see the data they held about my children. They directed me to ask my children’s school instead and to ask for their educational record. The difficulty with that is, it’s a different dataset.

My school is not the data controller of the National Pupil Database. I am not asking for a copy of my children’s educational records held by the school, but what information that the NPD holds about me and my children. One set of data may feed the other but they are separately managed. The NPD is the data controller for that data it holds and as such I believe has data controller responsibility for it, not the school they attend.

Why do I care? Well for starters, I want to know if the data are accurate.  And I want to know who else has access to it and for what purposes – school can’t tell me that. They certainly couldn’t two months ago, as they had no idea the NPD existed.

I went on to ask the DfE for a copy of the publicly accessible subject access request (SAR) policy and procedures, aware that I was asking on behalf of my children. I couldn’t find any guidance, so asked for the SAR policy. They helpfully provided some advice, but I was then told:

“The department does not have a publicly accessible standard SAR policy and procedures document.”  and “there is not an expectation that NPD data be made available for release in response to a SAR.”

It seems policies are inconsistent. For this other DfE project, there is information about the database, how participants can opt out and  respecting your choice. On the DfE website a Personal Information Charter sets out “what you can expect when we ask for and hold your personal information.”

It says: “Under the terms of the Data Protection Act 1998, you’re entitled to ask us:

  • if we’re processing your personal data
  • to give you a description of the data we hold about you, the reasons why we’re holding it and any recipient we may disclose it to (eg Ofsted)
  • for a copy of your personal data and any details of its source

You’re also entitled to ask us to change the information we hold about you, if it is wrong.

To ask to see your personal data (make a ‘subject access request’), or to ask for clarification about our processing of your personal data, contact us via the question option on our contact form and select ‘other’.”

So I did. But it seems while it applies to that project,  Subject Access Request is not to apply to the data they hold in the NPD. And they finally rejected my request last week, stating it is exempt:

SAR_reject

I appealed the decision on the basis that the section 33 Data Protection Act criteria given, are not met:

“the data subject was made fully aware of the use(s) of their personal data (in the form of a privacy notice)”

But it remains rejected.

It seems incomprehensible that third parties can access my children’s data and I can’t even check to see if it is correct.

While acknowledging section 7 of the Data Protection Act 1998 (DPA) “an individual has the right to ask an organisation to provide them with information they hold which identifies them and, in certain circumstances, a parent can make such a request on behalf of a child” they refused citing the Research, History and Statistics exemption (i.e. section 33(4) of the DPA).

Fair processing, another F for failure and F for attitude

The Department of Education response to me said that it “makes it clear what information is held, why it is held, the uses made of it by DfE and its partners and publishes a statement on its website setting this out. Schools also inform parents and pupils of how the data is used through privacy notices.”

I have told the DfE the process does not work. The DfE / NPD web instructions do not reach parents. Even if they did, information is thoroughly inadequate and either deliberately hides or does so by omission, the commercial third party use of data.

The Department for Education made a web update on 03/07/2015 with privacy information to be made available to parents by schools: http://t.co/PwjN1cwe6r

Despite this update this year, it is inadequate on two counts. In content and communication.

To claim as they did in response to me that: “The Department makes it clear to children and their parents what information is held about pupils and how it is processed, through a statement on its website,” lacks any logic.

Updating their national web page doesn’t create a thorough communications process or engage anyone who does not know about it to start with.

Secondly, the new privacy policy is inadequate in it content and utterly confusing. What does this statement mean, is there now some sort of opt out on offer? I doubt it, but it is unclear:

“A parent/guardian can ask that no information apart from their child’s name, address and date of birth be passed to [insert name of local authority or the provider of Youth Support Services in your area] by informing [insert name of school administrator]. This right is transferred to the child once he/she reaches the age 16. For more information about services for young people, please go to our local authority website [insert link].” [updated privacy statement, July 3, 2015]

Information that I don’t know exists, about a database I don’t know exists, that my school does not know exists, they believe meets fair processing through a statement on its own website?

Appropriate at this time of year,  I have to ask, “you cannot be serious?”

Fair processing means transparently sharing the purpose or purposes for which you intend to process the information, not hiding some of the users through careful wording.

It thereby fails to legally meet the first data protection principle. as parents are not informed at all, never mind fully of further secondary uses.

As a parent, when I register my child for school, I of course expect that some personal details must be captured to administer their education.

There must be data shared to adequately administer, best serve, understand, and sometimes protect our children.  And bona fide research is in the public interest.

However I have been surprised in the last year to find that firstly, I can’t ask what is stored on my own children and that secondly, a wide range of sensitive data are shared through the Department of Education with third parties.

Some of these potential third parties don’t meet research criteria in my understanding of what a ‘researcher’ should be. Journalists? the MOD?

The DfE web page which schools may link to from their data privacy statements states:

“We only hold information provided to us by third parties such as schools, local authorities and awarding bodies. We do not generate any new information about individuals, and the data provided to us is only used to make decisions about educational policy on an aggregate statistical basis.”  << this is clearly untrue when shared with commercial third parties are not involved in policy decisions.

To improve, there would be little additional time or work burden required to provide proper fair processing as a starting point, but to do so, the department can’t only update a policy on its website and think it’s adequate. And the newly updated suggested text for pupils is only going to add confusion.

The privacy policy text needs carefully reworded in human not civil service speak.

It must not omit [as it does now] the full range of potential users.

After all the Data Protection principles state that: “If you wish to use or disclose personal data for a purpose that was not contemplated at the time of collection (and therefore not specified in a privacy notice), you have to consider whether this will be fair.”

Now that it must be obvious to DfE that it is not the best way to carry on, why would they choose NOT to do better? Our children deserve better.

What would better look like? See part 3. The National Pupil Database end of year report: a D in transparency, C minus in security.

*****

[PS: I believe the Freedom of Information Officer tried their best and was professional and polite in our email exchanges, B+. Can’t award an A as I didn’t get any information from my requests. Thank you to them for their effort.]

*****

Updated on Sunday 19th July to include the criteria of my SAR rejection.

1. Our children’s school data: an end of year report card
2. The National Pupil Database end of year report: an F in fair processing
3. The National Pupil Database end of year report: a D in transparency, C minus in security

References:

[1] The National Pupil Database user guide: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/261189/NPD_User_Guide.pdf

[2] The Department for Education has specific legal powers to collect pupil, child and workforce data held by schools, local authorities and awarding bodies under section 114 of the Education Act 2005, section 537A of the Education Act 1996, and section 83 of the Children Act 1989. The submission of the school census returns, including a set of named pupil records, is a statutory requirement on schools under Section 537A of the Education Act 1996.

[3] Data tables to see the individual level data items stored and shared (by tabs on the bottom of the file) https://www.gov.uk/government/publications/national-pupil-database-user-guide-andsupporting-information

[4] The table to show who has bought or received data and for what purpose https://www.gov.uk/government/publications/national-pupil-database-requests-received

[5] Data Trust Deficit – from the RSS: http://www.statslife.org.uk/news/1672-new-rss-research-finds-data-trust-deficit-with-lessons-for-policymakers

[6] Talk by Phil Booth and Terri Dowty: http://www.infiniteideasmachine.com/2013/04/terris-and-my-talk-on-the-national-pupil-database-at-the-open-data-institute/

[7] Presentation given by Paul Sinclair of the Department for Education at the Workshop on Evaluating the Impact of Youth Programmes, 3rd June 2013
http://www.ifs.org.uk/docs/cayt_workshop_sinclair.pdf

[8] Source: [updated July 2015] The National Pupil Database User Guide: https://www.gov.uk/government/publications/national-pupil-database-user-guide-and-supporting-information

“The department has to make it clear to children and their parents what information is held about them, why it is held, the uses made of it by DfE and its partners. The department publishes a statement on its website setting out how the department processes pupils’ and children’s data.”

[9] The majority of schools’ policies I have reviewed online to date, don’t mention the Department for Education onwardly sharing data at all: example: http://www.leshigh.org/wp-content/uploads/2011/11/Parents-Handbook-2015.pdf

National Pupil Database, privacy, pupils, scope creep

The National Pupil Database end of year report: D for transparency, C minus in security.

Transparency and oversight of how things are administered are simple ways that the public can both understand and trust that things run as we expect.

For the National Pupil Database, parents might be surprised, as I was about some of the current practices.

The scope of use and who could access the National Pupil Database was changed in 2012 and although I had three children at school at that time and heard nothing about it, nor did I read it in the papers. (Hah – time to read the papers?)  So I absolutely agree with Owen Boswara’s post when he wrote:

“There appears to have been no concerted effort to bring the consultation or the NPD initiative to the attention of parents or pupils (i.e. the data subjects themselves). This is a quote from one of the parents who did respond:

“I am shocked and appalled that I wasn’t notified about this consultation through my child’s school – I read about it on Twitter of all things. A letter should have gone to every single parent explaining the proposals and how to respond to this consultation.”

(Now imagine that sentiment amplified via Mumsnet …)”
[July 2013, blog by O. Boswara]

As Owen wrote,  imagine that sentiment amplified via Mumsnet indeed.

Here’s where third parties can apply and here’s a list of who has been given data from the National Pupil Database . (It’s only been updated twice in 18 months. The most recent of which has been since I’ve asked about it, in .) The tier groups 1-4 are explained here on p.18, where 1 is the most sensitive identifiable classification.

The consultation suggested in 2012 that the changes could be an “effective engine of economic growth, social wellbeing, political accountability and public service improvement.”.  

Has this been measured at all if the justification given has begun to be achieved? Often research can take a long time and implementing any changes as a result, more time. But perhaps there has been some measure of public benefit already begun to be accrued?

The release panel would one hope, have begun to track this. [update: DfE confirmed August 20th they do not track benefits, nor have ever done any audit of recipients]

And in parallel what oversight governs checks and balances to make sure that the drive for the ‘engine of economic growth’ remembers to treat these data as knowledge about our children?

Is there that level of oversight from application to benefits measurement?

Is there adequate assessment of privacy impact and ethics in applications?

Why the National Pupil Database troubles me, is not the data it contains per se, but the lack of child/guardian involvement, lack of accountable oversight how it is managed and full transparency around who it is used by and its processes.

Some practical steps forward

Taken now, steps could resolve some of these issues and avoid the risk of them becoming future issues of concern.

The first being thorough fair processing, as I covered in my previous post.

The submission of the school census returns, including a set of named pupil records, has been a statutory requirement on schools since the Education Act 1996. That’s almost twenty years ago in the pre-mainstream internet age.

The Department must now shape up its current governance practices in its capacity as the data processor and controller of the National Pupil Database, to be fit for the 21st century.

Ignoring current weaknesses, actively accepts an ever-increasing reputational risk for the Department, schools, other data sharing bodies or those who link to the data and its bona fide research users. If people lose trust in data uses, they won’t share at all and the quality of data will suffer, bad for functional admin of the state and individual, but also for the public good.

That concerns me also wearing my hat as a lay member on the ADRN panel because it’s important that the public trusts our data is looked after wisely so that research can continue to use it for advances in health and social science and all sorts of areas of knowledge to improve our understanding of society and make it better.

Who decides who gets my kids data, even if I can’t?

A Data Management Advisory Panel (DMAP) considers applications for only some of the applications, tier 1 data requests. Those are the most, but not the only applications for access to sensitive data.

“When you make a request for NPD data it will be considered for approval by the Education Data Division (EDD) with the exception of tier 1 data requests, which will be assessed by the department’s Data Management Advisory Panel. The EDD will inform you of the outcome of the decision.”

Where is governance transparency?

What is the make up of both the Data Management Advisory Panel and and the Education Data Division (EDD)? Who sits on them and how are they selected? Do they document their conflicts of interest for each application? For how long are they appointed and under what selection criteria?

Where is decision outcome transparency?

The outcome of the decision should be documented and published. However, the list has been updated only twice since its inception in 2012. Once was December 2013, and the most recently was, ahem, May 18 2015. After considerable prodding. There should be a regular timetable, with responsible owner and a depth of insight into its decision making.

Where is transparency over decision making to approve or reject requests?

Do privacy impact assessments and ethics reviews play any role in their application and if so, how are they assessed and by whom?

How are those sensitive and confidential data stored and governed?

The weakest link in any system is often said to be human error. Users of the NPD data vary from other government departments to “Mom and Pop” small home businesses, selling schools’ business intelligence and benchmarking.

So how secure are our children’s data really, and once the data have left the Department database, how are they treated? Does lots of form filling and emailed data with a personal password ensure good practice, or simply provide barriers to slow down the legitimate applications process?

What happens to data that are no longer required for the given project? Are they properly deleted and what audits have ever been carried out to ensure that?

The National Pupil Database end of year report: a C- in security

The volume of data that can be processed now at speed is incomparable with 1996, and even 2012 when the current processes were set up. The opportunities and risks in cyber security have also moved on.

Surely the Department for Education should take responsibility seriously to treat our children’s personal data and sensitive records equally as well as the HSCIC now intends to manage health data?

Processing administrative or linked data in an environment with layered physical security (e.g. a secure perimeter, CCTV, security guarding or a locked room without remote connection such as internet access) is good practice. And reduces the risk of silly, human error. Or  simple theft.

Is giving out chunks of raw data by email, with reams of paperwork as its approval ‘safeguards’ really fit for the 21st century and beyond?

tiers

Twenty years on from the conception of the National Pupil Database, it is time to treat the personal data of our future adult citizens with the respect it deserves and we expect of best-in-class data management.

It should be as safe and secure as we treat other sensitive government data, and lessons could be learned from the FARR, ADRN and HSCIC safe settings.

Back to school – more securely, with public understanding and transparency

Understanding how that all works, how technology and people, data sharing and privacy, data security and trust all tie together is fundamental to understanding the internet. When administrations take our data, they take on responsibilities for some of our participation in dot.everyone that the state is so keen for us all to take part in. Many of our kids will live in the world which is the internet of things.  Not getting that, is to not understand the Internet.

And to reiterate some of why that matters, I go back to my previous post in which I quoted Martha Lane Fox recently and the late Aaron Swartz when he said: “It’s not OK not understand the internet, anymore”.

While the Department of Education has turned down my subject access request to find out what the National Pupil Database stores on my own children, it matters too much to brush the issues aside, as only important for me. About 700,000 children are born each year and will added to this database every academic year. None ever get deleted.

Parents can, and must ask that it is delivered to the highest standards of fair processing, transparency, oversight and security. I’m certainly going to.

It’s going to be Back to School in September, and those annual privacy notices, all too soon.

*****

1. The National Pupil Database end of year report card

2. The National Pupil Database end of year report: an F in fair processing
3. The National Pupil Database end of year report: a D in transparency

References:

[1] The National Pupil Database user guide: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/261189/NPD_User_Guide.pdf

[2] Data tables to see the individual level data items stored and shared (by tabs on the bottom of the file) https://www.gov.uk/government/publications/national-pupil-database-user-guide-andsupporting-information

[3] The table to show who has bought or received data and for what purpose https://www.gov.uk/government/publications/national-pupil-database-requests-received

[4] Data Trust Deficit – from the RSS: http://www.statslife.org.uk/news/1672-new-rss-research-finds-data-trust-deficit-with-lessons-for-policymakers

[5] Talk by Phil Booth and Terri Dowty: http://www.infiniteideasmachine.com/2013/04/terris-and-my-talk-on-the-national-pupil-database-at-the-open-data-institute/

[6] Presentation given by Paul Sinclair of the Department for Education at the Workshop on Evaluating the Impact of Youth Programmes, 3rd June 2013
http://www.ifs.org.uk/docs/cayt_workshop_sinclair.pdf

What is in the database?

The Schools Census dataset contains approximately eight million records incrementally every year (starting in 1996) and includes variables on the pupil’s home postcode, gender, age, ethnicity, special educational needs (SEN), free school meals eligibility, and schooling history. It covers pupils in state-funded primary, secondary, nursery, special schools and pupil referral units. Schools that are entirely privately funded are not included.

Pupils can be tracked across schools. Pupils can now be followed throughout their school careers. And it provides a very rich set of data on school characteristics. There is further use by linking the data from other related datasets such as those on higher education, neighbourhoods and teachers in schools.

Data stored include the full range of personal and sensitive data from name, date of birth and address, through SEN and disability needs. (Detail of content is here.)  To see what is in it download the excel sheet : NPD Requests.

 

The Department for Education has specific legal powers to collect pupil, child and workforce data held by schools, local authorities and awarding bodies under section 114 of the Education Act 2005, section 537A of the Education Act 1996, and section 83 of the Children Act 1989. The submission of the school census returns, including a set of named pupil records, is a statutory requirement on schools under Section 537A of the Education Act 1996.

National Pupil Database, politics, pupils, scope creep

Our children’s school data: an end-of-year report card

To quote the late Aaron Swartz: “It’s not OK to not understand the internet, anymore.”

Parents and guardians are trying their best.We leave work early and hurry to attend meetings on internet safety. We get told how vital it is that children not give away their name, age or address to strangers on social media. We read the magazines that come home in book bags about sharing their identity with players in interactive games.  We may sign school policies to opt out of permission for sharing photos from school performances on the school website.

And yet most guardians appear unaware that our children’s confidential, sensitive and basic personal data are being handed out to third parties by the Department of Education, without our knowledge or clear and accessible public accountability.

Data are extracted by the Department for Education [DfE] from schools, stored in a National Pupil Database [NPD], and onwardly shared.

Fine you may say. That makes sense, it’s the Department for Education.

But did you expect that the Ministry of Defence or Schools comparison websites may request or get given access to our children’s individual records, the data [detailed in the ‘NPD data tables’] that we provide to schools for routine administration?

School heads, governors, and every parent I have spoken with in my area, are totally unaware that data extracted by the Department of Education are used in this way.

All are surprised.

Some are shocked at the extent of data sharing at such an identifiable and sensitive level, without school and parental knowledge.

The DfE manages the NPD and holds responsibility to ensure we know all about it. But they’re not ensuring that pupils and parents are told before the data extraction, who else gets access to it and for what purposes. That fails to process data fairly which is a requirement to make its use lawful.

There’s no way to opt out, to check its accuracy or recourse for anything incorrect.

As our lives involve the ever more advanced connectivity of devices, systems, and services, that’s simply not good enough. It’s not a system fit for the 21st century or our children’s digital future.

While the majority of requestors seem to access data for bona fide research in the public interest, some use it for bench marking, others are commercial users.

Is that what pupils and parents expect their data are used for?

And what happens in future when, not if, the Department chooses to change who uses it and why.

How will we know about that? Because it has done so already.

When school census data first began, it extracted no names. That changed. Every pupil’s name is now recorded along with a growing range of information.

Where it began with schools, it is now extended to nursery schools; childminders, private nurseries and playgroups.

Where it was once used only for state administrative purposes, since 2012 it has been given to third parties.

What’s next?

Data should be used in the public interest and must be shared to adequately administer, best serve, understand, and sometimes protect our children.

I want to see our children’s use of technology, and their data created in schools used well in research that will enable inclusive, measurable benefits in education and well being.

However this can only be done with proper application of law, future-proofed security, and respectful recognition of public opinion.

The next academic year must bring these systems into the 21st century to safeguard both our children and the benefits that using data wisely can bring.

Out of sight, out of date, out of touch?

The data sharing is made possible through a so-called ‘legal gateway’, law that gives permission to the Secretary of State for Education to require data from schools.

In this case, it is founded on legislation almost twenty years old.

Law founded in the 1996 Education Act and other later regulations changed in 2009 give information-sharing powers to the Secretary of State and to public bodies through law pre-dating wide use of the Internet, social media, and the machine learning and computer processing power of today.

Current law and policies have not kept pace with modern technology. 2015 is a world away even from 2009 when Pluto was still a planet.

Our children’s data is valuable, and gives insights into society that researchers should of course use to learn from and make policy recommendations. That has widespread public support in the public interest. But it has to be done in an appropriate and secure way, and as soon as it’s for commercial use. there are more concerns and questions to ask.

As an example why NPD doesn’t do this as I feel it should, the data are still given away to users in their own offices rather than properly and securely accessed in a safe-setting, as bona fide accredited researchers at the Office of National Statistics do.

In addition to leaving our children’s personal data vulnerable to cybersecurity threats, it actively invites greater exposure to human error.

Remember those HMRC child benefit discs lost in the post with personal and bank data of 25 million individuals?

Harder to do if you only access sensitive data in a safe setting where you can walk out with your research but not raw files.

When biometrics data are already widely used in schools and are quite literally, our children’s passport to the world, poor data management approaches from government in health and education are simply not good enough anymore.

It’s not OK anymore.

Our children’s personal data is too valuable to lose control of as their digital footprint will become not an add-on, but integral to everything they do in future.

Guardians do their best to bring up children as digitally responsible citizens and that must be supported, not undermined by state practices.

Children will see the divide between online and ‘real’-life activities blend ever more seamlessly.

We cannot predict how their digital identity will become used in their adult lives.

If people don’t know users have data about them, how can we be sure they are using it properly for only the right reasons or try and repair damage when they have not?

People decide to withhold identities or data online if they don’t trust how they will be used, and who will use it well.

Research, reports and decision making are flawed if data quality is poor. That is not in the public interest.

The government must at least take responsibility for current policies to ensure our children’s rights are met in practice.

People who say data privacy does not matter, seem to lack any vision of its value.

Did you think that a social media site would ever try to control its users emotions and influence their decision-making based on the data they entered or read? It just did.

Did you foresee five years ago that a fingerprint could unlock your phone? It just did.

Did you believe 5 months ago the same fingerprint accessible phone would become an accepted payment card in England? It just did.

There is often a correlation between verification of identity and payment.

Fingerprinting for payment and library management has become common in UK schools and many parents do not know that parental consent is a legal requirement.

In reality, it’s not always enacted by schools.

Guardians can find non-participation is discouraged and worry their child will be stigmatised as the exception.

Yet no one would seriously consider asking guardians to give canteens their bank card PIN.

The broad points of use where data are created and shared about our children mean parents can often not know who knows what about them.

What will that mean for them as adults much of whose lives will be digital?

What free choice remains for people who want to be cautious with their digital identities? 

Many systems increasingly require registration, some including biometric data, sometimes from vulnerable people, and the service on offer is otherwise denied.

Is that OK anymore? Or is denial-of-service a form of coercion?

The current model of state data sharing often totally ignores that the children and young people whose personal data are held in these systems are not asked, informed or consulted about changes.

While Ministers talk about wanting our children to become digital leaders of tomorrow, policies of today promote future adults ill-educated in their own internet safety and personal data sharing practices.

But it’s not OK not to understand the internet anymore.

Where is the voice of our young people talking about who shares their information, how it is used online, and why?

When shall we stop to ask collectively, how personal is too personal?

Is analysing the exact onscreen eye movement of a child appropriate or invasive?

These deeply personal uses of our young people’s information raise ethical questions about others’ influence over their decision making.

Where do we draw the line?

Where will we say, it’s not OK anymore?

Do we trust that all uses are for bona fide reasons and not ask to find out why?

Using our children’s data across a range of practices in education seem a free for all to commercially exploit, with too little oversight and no visibility of decision making processes for the public,whose personal data they profit from.

Who has oversight for the ethical use of listening software tools in classrooms, especially if used to support government initiatives like Channel in ‘Prevent’?

What corrective action is taken if our children’s data are exposed through software brought into school over which parents have no control?

The policies and tools used to manage our children’s data in and outside schools seem often out of step with current best-in-class data protection and security practices.

Pupils and parents find it hard to track who has their personal data and why.

While the Department for Education says what it expects of others, it appears less committed to meeting its own responsibilities: “We have been clear that schools are expected to ensure that sensitive pupil information is held securely. The Data Protection Act of 1998 is clear what standards schools are expected to adhere to and we provide guidance on this.” 

A post on a webpage is hardly guidance fit to future proof the data and digital identities of a whole generation.

I believe we should encourage greater use of this administrative data for bona fide research. Promoting broader use of aggregated and open data could also be beneficial. In order to do both, key things should happen that will make researchers less risk averse in its use, and put data at reduced risk of accidental or deliberate misuse by other third parties. Parents and pupils could become more confident that their data is used for all the right reasons.

The frameworks of fair processing, physical data security, of transparent governance and publicly accountable oversight need redesigned and strengthened.

Not only for data collection, but its central management, especially on a scale as large as the National Pupil Database.

“It’s not OK not to understand the internet anymore.”

In fact, it never was.

The next academic year must bring these systems into the 21st century to safeguard both our children and the benefits that using data wisely can bring.

The Department for Education “must try harder” and must start now.

********

If you have questions or concerns about the National Pupil Database or your own experience, or your child’s data used in schools, please feel free to get in touch, and let’s see if we can make this better. [Email me as listed above right.]

1. An overview: an end of year report on our Children’s School Records
2. The National Pupil Database end of year report: an F in fair processing
3. The National Pupil Database end of year report: a D in transparency, C- in security

********

References:

[1] The National Pupil Database user guide: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/261189/NPD_User_Guide.pdf

[2] Data tables to see the individual level data items stored and shared (by tabs on the bottom of the file) https://www.gov.uk/government/publications/national-pupil-database-user-guide-andsupporting-information

[3] The table to show who has applied for and received data and for what purpose https://www.gov.uk/government/publications/national-pupil-database-requests-received

[4] Data Trust Deficit – from the RSS: http://www.statslife.org.uk/news/1672-new-rss-research-finds-data-trust-deficit-with-lessons-for-policymakers

[5] Talk by Phil Booth and Terri Dowty: http://www.infiniteideasmachine.com/2013/04/terris-and-my-talk-on-the-national-pupil-database-at-the-open-data-institute/

[6] On 1 September 2013 sections 26 and 27 of the Protection of Freedoms Act 2012 came into force, requiring schools to seek parental consent before collecting biometric data, such as fingerprints.

Thinking to some purpose