Dear Department for Work and Pensions,

On Wednesday 3 April 2013, Mr John Mackintosh of Atos published a letter in The Guardian in which he stated “in only 0.3% of cases was a successful appeal due to an error in an Atos Healthcare assessment”.

1) Precisely how did he calculate this figure?

2) What sources of information/data did he use and what steps did he take to validate its authenticity?
[If it has been in any way derived, I would like to see the exact arithmetic behind it]

3) If he is correct, the probability of two consecutive errors on the same claimant would be about 1 in 100,000 and for three consecutive errors would be about 1 in 37 million. Do the stats on “multiple” errors bear these rates out?

4) What cumulative information does Atos/DWP hold on claimants who have been through a WCA on more than one occasion and does it consider this history when making an assessment/decision?

Yours faithfully,

Rose Cullingford

Dear Department for Work and Pensions,

Please pass this on to the person who conducts Freedom of Information reviews.

I am writing to request an internal review of Department for Work and Pensions's handling of my FOI request 'Atos error rates'.

Firstly, as regards Q1/2, there has been much publicity recently about public bodies generally and DWP in particular misrepresenting statistical information to suit its own ends and the same applies to Atos here.

1) Yes, 0.3% is quoted in the report, but it is plainly 0.3% of 64.3% (not 100% as implied) = 0.5%.

2) This 0.3/0.5% relates to ERRORS, not bad judgement and the overall quality of the assessment, which are contained within two other areas:

a) The 40.5% where “cogent oral evidence” was provided by the claimant. If the claimant was able to do this under the pressure of a Tribunal hearing, clearly they would have done it at the WCA too.
b) The 35.7% where no reasons were attached.
To quote 0.3% without any of these qualifications is a perfect example of what the controversy is about. No response required.

As regards the other questions, I would like more specific detail:

Q3: This is not asking for explanations, opinions or confirmation, it is asking about information available on claimants who have experience wrong decisions on more than one occasion. Where is it available?

As one would expect that having made the first mistake, both DWP & Atos would make extra efforts to ensure they do not make the same mistake twice, please also provide any information you have that shows you both do thnis to avoid repeating errors.

Q4: You refer to “available evidence provided in the referral” – what could this include? Is there as standard "pack"?

Please confirm specifically what “all data Atos Healthcare has” includes. I am particularly interested to see if Atos makes any attempt to learn from its previous errors.

A full history of my FOI request and all correspondence is available on the Internet at this address:
http://www.whatdotheyknow.com/request/at...

Yours faithfully,

Rose Cullingford

Dear DWP DWP Medical Services Correspondence,

Thank you for the IRR, but whether I like them or not, I do understand the limitations of FoI legislation and would have been happy if you’d have said clearly at the outset that the information does not exist in which case the IRR would not have been necessary – you may regard this as implied in your first response, but it is not actually stated.

As an aside, it is extremely disappointing and telling that you do not track claimants who have suffered multiple errors – making a mistake with someone once is bad enough, repeating the mistake is unforgivable. How on earth can you claim an overall improvement to the process if you do not even collate obviously worthwhile comparisons such as this?

Yours sincerely,

Rose Cullingford