Audit Committee, 05 Dec 2007

Meeting date: Wednesday, December 5, 2007

Official Report 162KB pdf

---

“Overseas staff in the NHS— pre-employment checks”

Item 3 is consideration of the Auditor General for Scotland's report, "Overseas staff in the NHS—pre-employment checks". I invite the Auditor General to make some introductory remarks.

This short report examines whether national health service boards are following their own policies on pre-employment checks of staff who are recruited from outside the European economic area, which is made up of European Union countries, Iceland, Norway and Liechtenstein and includes Swiss nationals.

Members might recall that this review does not form part of Audit Scotland's planned programme of national performance studies. Following this summer's security incidents in London and Glasgow, in which overseas staff in the NHS were said to be involved, the Cabinet Secretary for Health and Wellbeing requested a review of whether the NHS in Scotland is in line with the Centre for the Protection of National Infrastructure's guidelines. The Scottish Government also reviewed whether its national policy and guidance on pre-employment screening for overseas staff complied with the same guidelines.

The Government asked me whether Audit Scotland would review whether NHS boards are following their own pre-employment screening procedures. In recognition of the public interest in the area, I agreed to undertake the work. However, I should emphasise that I set the terms of reference for our review and made it clear that I wished to report my findings in public. That is what I am doing today.

All of us, including me and my Audit Scotland colleagues, recognise the important role that overseas health care staff play in supporting the NHS in the provision of care to all the people of Scotland. To ensure that it recruits the right person for the job, the NHS carries out pre-employment screening for all staff, including checking whether they have the necessary qualifications and experience. Boards are also required to ensure that overseas staff have the right permissions to work in the United Kingdom. The guidance on pre-employment checks is limited to assessing an individual's suitability for and capability to do the job and does not cover, for example, ideological beliefs.

At this point, with the committee's agreement, I invite Barbara Hurst briefly to outline the report's key findings.

When we asked all 14 boards to provide us with the number of overseas staff that they employ, they reported that, at September 2007, there was a total of 1,161 overseas staff in employment, 89 per cent of whom were doctors or nurses. However, that figure is likely to be an underestimate, as we found that it was difficult for boards to provide us with accurate numbers, particularly for staff with the immigration status of indefinite leave to remain. Moreover, we found that not all staff information was held centrally, which made it more difficult to collate.

We examined compliance with the pre-employment screening procedures in five boards: Ayrshire and Arran NHS Board; Grampian NHS Board; Greater Glasgow and Clyde NHS Board; Lanarkshire NHS Board; and Lothian NHS Board. We selected those boards because information that had been supplied previously showed that they had the highest numbers of overseas staff. We looked in a sample of files in each of the boards for evidence of checks on identity; reference checks; qualification and regulatory checks, such as General Medical Council checks for doctors; checks on work and residence permits; and Disclosure Scotland checks, which examine criminal history in Scotland.

Given the problems that boards had in giving us accurate numbers for overseas staff, we cannot give an absolute assurance that the sample was representative, so our findings are indicative. However, we found evidence of a high rate of compliance with the checks. We reviewed 235 records in total, and the percentage of files containing evidence of the checks ranged from 90 to 99 per cent, depending on the check in question. It is fair to say that we were looking to see that the evidence was on the file. The fact that the evidence was not on the file does not necessarily mean that the check was not carried out—it might not have been recorded.

Our report contains a number of recommendations for the Government and NHS boards, including the need to make smarter use of the national workforce information system and more proactive use of immigration status fields. We also recommend that boards prioritise the wider use of their electronic records to support better management information on staff in general.

As the Auditor General said, the Government is revising its policy and guidance on pre-employment checks for overseas staff, which we understand will be issued this month. We think that that will include guidance on overseas criminal history checks.

On the boards that were not subject to our sampling, we are asking our auditors to check that there are similar processes throughout Scotland.

We are happy to answer questions.

Thank you for that introduction. This important and timely report arises from the terrorist attack on Glasgow airport and the concern thereafter about the employment of overseas nationals in the health service and other public services. The information that the report provides is positive and encouraging, but there are still areas of concern.

Page 3 of the report says that a total of 235 personnel files of overseas staff were examined. How were those files selected? Was the selection random or were particular individuals picked out?

The files were selected at random, but given that we did not know the total number of overseas staff, the selection might have been skewed. That is the reason for the slight caveat about whether the sample was representative.

I want to pick up the point about those with indefinite leave to remain. Is there a further problem to be looked at there?

Is information on tracking the movement of staff between health boards, or into other forms of employment, contained in the study? Is there such a tracking system?

On staff with indefinite leave to remain, I will be modest and say that we do not claim to be immigration specialists. When we started this work, we found that immigration was pretty complicated. As an employer, a board might not necessarily need to know whether someone has been granted indefinite leave to remain. The checks that we were looking at were checks on work permits, for example, which people with indefinite leave to remain do not need.

However, our view was that boards should know which of their staff were in that category, if only for workforce planning purposes. They would need to know all the migration flows into and out of their workforce. We are not saying that this is necessarily a risk factor. We reviewed some files of staff with indefinite leave to remain, from which work permit checks were excluded. Of course, every member of staff has to have their references and identity checked.

Tracking staff was not part of what we were looking at. We were doing a focused piece of work on individual files. However, one would expect the same checks to be made if a member of staff moved between board areas. We did not specifically consider the tracking of staff, but we would have done the same review if we had any such staff in our sample.

Page 11 of the report states:

"We were able to identify evidence of reference checks being completed in 90 per cent of cases"

only. Will you comment on that? Does that mean that 10 per cent of references—116 or so—were not checked?

Ideally, one would want the files to show that 100 per cent of references had been checked, irrespective of whether a member of staff was from overseas. However, we are not saying that a reference was not checked just because there was no evidence of that on file. That was the difficulty. The focus was so much on the evidence on the files that we could not consider the whole recruitment process. However, you are right. Ninety per cent is the worst figure in the summary of compliance. We would certainly expect boards to ensure that they put evidence on their files.

You have stated that it is impossible to be precise about the number of overseas staff in the national health service. One reason that you gave for that is that some information is not held centrally. Is that common across all the health boards?

It was common in the boards that we reviewed. However, in light of the number of staff who are employed, it may be appropriate that files should be with line managers, in particular hospitals or whatever.

We thought that better use could be made of workforce information systems. Key bits of information could be held in a central place so that people could retrieve them. However, there is no reason why things could not be decentralised for practical operational purposes. Probably all boards have records in their various operational units, but we would like all boards to have much better and more comprehensive central information too. That should be perfectly possible if the information is in an electronic format.

I want to follow up on the question that Jim Hume asked. Is it not worrying that you could not access references in 10 per cent of cases?

I suppose that the answer to that is yes and no. I do not want to be an apologist, but the committee should realise that we looked at the evidence on the files. We cannot categorically say that the references in question were not checked. All that we can say is that 90 per cent of the files that we checked contained evidence of reference checks.

Did the other files simply contain boxes that had been ticked?

No. The exercise found a mixed range of record keeping systems. Catherine Vallely may be able to give more details about that. Some files contained comprehensive copies of everything, whereas others contained copies of some papers with ticked checklists that showed what had been checked that was not kept on the file.

Could you follow the information back? Was it evidenced? Had people signed papers after ticking boxes? Could you go to the person who had ticked a box and ask them whether they had seen the reference?

Yes. In some cases, there was a pro forma with a tick from somebody in human resources who had completed and signed it, and we raised any queries with them.

Paragraph 26 of the report is even more worrying. It states:

"There was evidence of work permits for 96 per cent of staff".

That implies that there was no evidence that 4 per cent of those staff had work permits.

There were no photocopies of the work permits on file, but that does not mean that those staff did not have work permits. There was inconsistent record keeping.

You have recommended that, in future, all boards should keep copies of references and work permits and not rely on tick boxes.

Yes.

Good.

We have also recommended that people should streamline trigger processes to ensure that work permits are kept up to date. At the moment, the system relies on a range of people in boards to do that.

I hope that the report is followed up. What will happen now? I keep asking that question in Audit Committee meetings. Audit Scotland produces excellent reports, and we must ensure that they are followed up.

To be fair, that job is not for Audit Scotland but for us, as a committee. We will decide, at a later stage, what action to take to pursue the matter.

Excellent.

Although we, too, will ensure that this work is carried out in every board, not just in the five sample boards.

On reference checks, even if ancillary and other workers were not all checked, did you identify that doctors were at least given reference checks?

In a sense, the record keeping for doctors is likely to be better, purely because they also have to have regulatory checks, such as checks by the GMC. Perhaps Catherine Vallely has more information on that.

I do not have a breakdown of the 10 per cent whose references were not checked, but I agree that the regulatory checks on doctors provide reassurance.

The "Qualifications and Regulatory" category shows a 99 per cent compliance rate.

On the indefinite leave to remain category, I noticed that the figures for NHS Fife and NHS Lanarkshire are quite high. Is that down to different recording practices?

The point is an interesting one. NHS Lanarkshire was one of the five sample boards. Its system for recording staff with indefinite leave to remain is one of the best that we saw. We are therefore not surprised that its returns show higher numbers in that category.

The information from NHS Lanarkshire helped to inform our suspicion that other boards employ people in that category but do not record the information because their systems do not enable them to retrieve the data.

You spoke of inconsistent record keeping. Do boards have the administrative machinery to impose an effective and up-to-date system? If so, can they do that within existing resources? What is the extent of the variation across boards?

There is variation across and within boards. When the health service introduced its electronic workforce information system—I cannot remember its exact name—a few years ago, the aim was to improve record keeping significantly. Record keeping, in terms of centrally held information, will probably improve over the next few years, because people who are employed over that period are far more likely to be registered on that system. That will mean that information will no longer have to be produced by going back through records retrospectively. In our view, the system can be improved within current resources. Good record keeping on individual staff members is not rocket science. Given that we are talking about staff who work with vulnerable people, we should expect good record keeping.

You spoke about overseas criminal record checks, which you said complement Disclosure Scotland checks. What is the timescale for those checks? Are they doable?

Obviously, this is an important area. In our work, we followed up what boards were doing, set against their local policies and procedures. Given that overseas criminal record checks did not feature highly in those policies and procedures, we did not include them as a pre-employment check. That said, we found evidence in some of the files that we reviewed that staff had had that check. The overseas criminal record check is probably the most difficult area of overseas staff pre-employment checking.

The national Criminal Records Bureau has details on the United Kingdom's bilateral agreements with other countries. Catherine Vallely will correct me if I am wrong, but I think that most of them are with Commonwealth or European Union countries. The bureau gives employers advice on conducting checks with other countries in that regard. That said, the process is, of course, a difficult one for employers.

The expectation of NHS employers is that they should undertake a risk assessment, because that will show whether such a check is worth while. Clearly, the check should absolutely be done for workers from countries with which we have an agreement.

I have a follow-up question. Paragraph 28 states that some of the personnel records had evidence of overseas checks. What sort of percentage are we talking about?

It is probably not very high. Catherine Vallely might be able to throw some light on that.

As Barbara Hurst mentioned, such checks were not one of the key issues that we examined, but the number was not significant at all.

I presume that those checks were done fairly randomly, so there was no particular pattern—perhaps they related to the type of job or health board.

There was no pattern.

Paragraph 27 states that there is evidence of Disclosure Scotland checks being carried out in 95 per cent of cases. Such checks are a legal requirement when people work with children. Is the issue with the remaining 5 per cent of cases simply that records were not kept properly?

There are special rules about how records from Disclosure Scotland are kept. I ask Catherine Vallely, who is becoming our resident expert, to explain what that actually means. It may explain why the figure is only 95 per cent. Can you help, Catherine?

Again, the issue is inconsistent record keeping. The Disclosure Scotland form is not kept on the file—there is a reference number. In some cases, the reference number was not noted on the checklist. That accounts for the 5 per cent figure. We had no real way of checking anything further, as nothing is kept on file, for data protection reasons.

I assume that agency staff are included in the figure of 1,161 overseas staff.

No, they are separate. The figure of just over 1,000 is for health board employees. Agency staff are another category. We will ask the auditors of the body that has the national contracts for agency staff to do a similar piece of work to that which we have done on the five sample boards.

Is there a timescale for that work?

We need to plan that in the audit programme, but I guess that it will happen during the current audit year, so I hope that it will be done by March.

Do you have an indication as to the number of agency staff who work in the NHS at present?

I thank the Auditor General, who just provided me with some figures as I looked at you blankly. We have figures for nurses, but not doctors. In 2006-07, 728 whole-time equivalent agency nurses were working in the NHS. If the committee is interested, we can provide more up-to-date information for all the categories. We could probably find out the information for doctors.

The report states:

"The guidance on pre-employment checks is limited to assessing an individual's suitability and capability to do the job and does not cover their ideological beliefs."

However, it continues:

"As employers … NHS bodies can develop local procedures to supplement the statutory requirements."

Was there any evidence of NHS boards introducing additional procedures?

The difficulty with the work was that the checks are basic good employment practice that organisations perform for most staff, although there are additional checks on work permits for overseas staff. We need to be absolutely clear: there is no way in which an NHS employer could spot a potential terrorist through that process. How could they? In the same way, an employer recruiting a UK citizen would not be able to spot a potential terrorist. Boards may put in place additional local checks to supplement the good-practice employment checks, but they would not be part of the activities around what is, in a sense, a legitimate Home Office UK immigration policy.

I have an entirely different question. Mr Black helpfully set out, as does the report, how we describe overseas staff, which is those who are not from the European Union, Iceland, Norway, Liechtenstein or Switzerland. Are residents of the Channel Islands and the Isle of Man considered as overseas staff?

This is absolutely dreadful; you are exposing our total lack of geographical knowledge about immigration. I genuinely do not know.

It is very interesting that you do not know. Do you know, Mr Black?

No.

Do you know, Ms Vallely?

No.

It is interesting that three very qualified, experienced people are unable to tell me. I can tell you that the Channel Islands are not part of the European Union. Are they part of the European economic area? Did you discuss that? Were any NHS staff from the Channel Islands or the Isle of Man? It is very interesting.

Thank you for exposing our ignorance.

The Channel Islands and the Isle of Man are Crown dependencies and are part of the British isles. I can tell you a lot more about them if you are interested.

We can leave that discussion to another day.

I think that I have made my point.

Indeed.

You have caught us bang to rights. The question that I anticipated—which has not yet been asked, but is perhaps to come—concerns the situation of Commonwealth citizens. Apparently, it is very complicated. Perhaps this issue is a subset of that.

I have a much more mundane question. Page 6 of the report gives a breakdown of the 1,161 members of NHS staff who are from overseas. Of the 45 staff who are categorised as "Other/unidentified immigration status", two thirds—a total of 30—work in NHS Fife. Was there anything unusual about the way in which the categories were allocated in NHS Fife?

It is difficult to say, because Fife NHS Board was not one of the boards that we examined. We may have more information once the auditor has looked at it. In a sense, at least the board was honest about the numbers. I am afraid that that is all I can say.

Going back to my previous question about how the figures break down into doctors and ancillary staff, I see that exhibit 3 on page 8 of the report shows that all the overseas staff in NHS Ayrshire and Arran were doctors. Appendix 2 on the last page of the report shows that there was no evidence of a Disclosure Scotland check on file for five of the 44 staff sampled. Also, for five of the 44 staff sampled there was no evidence of a check on file for work references. That seems quite a large percentage. I would have hoped that doctors above all would have to undergo such checks.

I can explain something here. We thought that it was unusual for 100 per cent of a board's overseas staff to be doctors. Ayrshire and Arran NHS Board was included in the sample, therefore we can say that some of the files that we sampled were not doctors.

That means that the chart is wrong.

The chart is based on the information that boards provided to us. Remember that I said earlier that they could not give us an accurate picture. We know that there are some overseas nurses in NHS Ayrshire and Arran because we have seen their files.

Knowing that, why did you not change the chart?

We asked all the boards to reconfirm the numbers.

So the numbers came from the boards. There is something wrong there.

Okay, I thank members for their questions. I remind members that we will return to the issue under item 5—item 6 on the amended agenda—when we will discuss how we intend to deal with the report. I thank the Auditor General and his team.