Audit Committee, 25 Nov 2003

Meeting date: Tuesday, November 25, 2003

Official Report 165KB pdf

---

“Catering for patients”

Item 2 is our consideration of Audit Scotland's report "Catering for patients". I invite the Auditor General for Scotland and his team to give the committee a brief outline of the report.

Thank you. With the convener's agreement, I invite Barbara Hurst to lead on this item.

At £55 million, expenditure on hospital catering services is relatively small in terms of the overall health budget. However, given that the service affects all hospital patients, we feel that it is important.

This is our first baseline review of Scotland's hospital catering services. We undertook a detailed review of quality, patient satisfaction and cost and management of the catering service at 41 hospitals in 26 national health service bodies. The review covered about 50 per cent of staffed beds. As NHS Quality Improvement Scotland was developing standards for food, fluid and nutrition at the same time as our review, we made a decision early in the process to include nutritional care only as a high-level issue. We did not want to duplicate what NHS Quality Improvement Scotland was doing.

Overall, the report is very positive about catering services. We found high levels of patient satisfaction: the average was 92 per cent. Even allowing for what my son calls a captive audience, that is pretty good going. However, we found four areas in which we think hospitals need to do more work. The first is nutritional care, which needs to be given a higher priority. Around 20 per cent of trusts do not have a tool for screening patients who are at risk of malnutrition and a further 10 per cent are using a tool that has not been validated. Most trusts analyse their menus for nutritional content, but the extent to which they do that varies considerably. There is room for some trusts to do more in that area.

Secondly, ward wastage due to unserved meals is high in a number of trusts. It ranged from 1 per cent to over 40 per cent. Although particular delivery systems such as bulk delivery can lead to a bigger risk of wastage, we feel that wastage can be managed at a local level. If a 10 per cent wastage target could be achieved across hospitals, up to £1.9 million could be saved and reinvested in the catering service. It is important to say that responsibility for that lies not only with catering departments; there is a real issue of communication between wards and the catering service, which could be improved.

Thirdly, we found a considerable variation in the daily spending on food and beverages for patients, ranging from £1.25 a day to just over £3. It is interesting that that variation is similar to the position in England, Wales and Northern Ireland, so it is not unusual. However, we think that it should be examined. We found that although two in five hospitals have set a daily food allowance for patients, less than half of those hospitals use that figure to set their catering budget. That leads us to ask why they set an allowance at all.

The final area that we thought needed examination was the subsidy of catering services. Three quarters of the hospitals that we reviewed subsidise catering services for staff and visitors and most of them are either unaware that they do so or unaware of the level of subsidy that they put in. That situation arises due to poor management information or because hospitals are following old guidance that needs to be reviewed. We suggest that such subsidy amounts to £4.2 million, with an average subsidy per hospital of about £110,000. That money could go back into catering for patients.

The report makes several recommendations for trusts and for the Scottish Executive Health Department and we intend to follow it up in about 18 months' time. We are in early discussions with NHS Quality Improvement Scotland about whether there should be a joint follow-up.

The committee will consider its approach to the report under agenda item 9, but I open the meeting to members to ask questions. Before Margaret Jamieson and George Lyon ask their questions, I slap myself on the wrist yet again because I forgot to ask people present to turn off their pagers and mobile phones.

I should perhaps declare an interest—it may be a long time since I worked in NHS catering departments, but I have read the report and it is amazing how quickly it all comes back to me.

Is there any evidence of differences in nutritional values and in the screening of patients between long-stay hospitals and acute hospitals?

On staff meals, the subsidy was introduced in 1978 to ensure that meals were available for staff, in particular for ancillary staff, not for those who were deemed to be higher earners. The subsidy has continued and successive Governments have obviously been happy with that. If we remove the subsidy, the lower paid will suffer and there will be a knock-on cost due to illness and staff turning up for work tired. We must consider the issue in context and not regard it as a simple way to save money.

The report does not include sufficient detail on the training of catering staff. It considers food handling and hygiene but does not consider people who are involved in the preparation and cooking of food and does not determine their qualifications or whether they are suitably qualified to carry out those tasks. Sorry.

No, that is okay. I will kick off with the subsidy issue. We were careful to say that the subsidy should be taken away without thinking about the reasons why it is there. For us, the issue is that trusts did not seem to be aware that they were subsidising staff meals. Even where they were aware of that, the subsidy was not targeted at lower-paid workers but was a blanket subsidy. I take Margaret Jamieson's point, and it is for local trusts to determine whether they want to continue to subsidise staff meals, but we would like there to be more transparency about how that is done.

We were careful to consider other issues besides nutritional content, particularly with regard to meals for older patients in the longer-stay wards. We looked not only at nutritional content, but at choice and menu rotation, to ensure that patients were getting the right sort of meals and that their food had the right content. I will pass you over to Angela Cullen, who knows the detail of the study. She will also be able to address the training issues.

On nutritional care, as Barbara Hurst said, we undertook a high-level review because NHS Quality Improvement Scotland was going to come in later to consider the issues. We asked whether the menus were nutritionally analysed and whether a nutrition screening tool was used. We have not highlighted this in the report, but the primary care trusts came out better on the question of the nutritional screening tool. As the acute trusts still have long-stay wards, that is an issue. There was no difference between the acute trusts and long-stay hospitals on the question of the nutritional analysis of menus.

We have not highlighted it in the report, but we asked auditors to look at the qualifications that catering staff in hospitals had. Similarly, we asked them to consider food safety and hygiene training. The auditors looked at the qualifications of a sample of staff to see whether they were commensurate with the jobs that they were doing. We found that there was no problem. People who were not qualified were doing lower-grade jobs but were being trained to become higher-grade staff.

One of the problem areas that we picked up on was the recruitment and retention of staff. It is difficult to get qualified staff to work in the NHS for the money that NHS staff are paid. However, there is no problem with the qualifications of the staff who are in place.

I would like you to expand on two issues. First, there seems to be a huge range of figures for wastage—between 1 per cent and 40 per cent. What is causing the differences? What did you identify as the key drivers in whether there is efficient use of food or 40 per cent wastage? Secondly, your submission states that the net cost of catering ranges between £3.50 and £7.50 per patient day and that food and beverage costs per patient day range between £1.25 and £3.03. Can you explain what is behind the differences in those figures? Is it volume related—bigger hospitals will clearly have lower food costs—or is rurality an issue? What did you find to be the common denominator in determining whether costs are high or low?

I shall pick up on wastage first. Two issues contribute to wastage, the main one being lack of communication between the wards and the catering departments. We found some examples of very simple good practice for ensuring that there is good communication between the two. We also found that, where communication is not so good, there is likely to be greater wastage.

The other issue is simple and concerns the size of the trays that are used in preparing the food. There are different ways of delivering food to the wards: either it can be plated or it can be delivered in trays from which it is served. In the latter scenario, there is likely to be more wastage, although that is not inevitable. Different sized trays can be used to avoid a one-tray-fits-all scenario.

Those are two simple issues. I agree that the range of wastage varies hugely, and if some hospitals are managing to have low wastage, all hospitals can do it.

For clarification, is it an issue among hospitals or among trusts? Do some trusts manage the situation better, or is it a matter of individual hospital management?

It is an issue for individual hospitals. The waste that we are talking about is the waste that is in the wards—the meals that go to the wards but are not taken by patients; it is not the waste that is left on the plates once the patients are finished or the wastage that is sitting in the catering departments.

It is the food that is prepared and never eaten.

It is only the food that is in the wards.

The other question was about—

It was about the range in costs. What is the explanation for that?

We share some of your uncertainty about why the range should be so huge. Clearly, matters such as portion sizes and waste have an impact. However, I am not sure that we bottomed exactly why there is such a range in costs.

We can tell you what is not a factor.

I suppose that that is a start.

The size of hospitals, for example, is not a factor.

So the range in costs has nothing to do with volume.

No, it is not about volume, although we had thought that that might be the case. Nor is it about the way in which food is prepared. We have tried to consider everything that we can to find the reason for cost differences, but we cannot identify any common factor, either in hospitals where costs are low or in hospitals where they are high.

Hospital catering divisions certainly have a reputation for driving an extremely hard bargain when they ask for deliveries—any meat supplier will tell you that. Hospital caterers are reputed to work to a set budget of 23p per meal, which has to cover the meat or whatever else is in the meal. Given the actual cost of the individual meal per patient, why should overall costs be so high?

As John Simmons says, we used just about every correlation that we could think of to try to understand the range in costs, but we could not find any strong relationships that could explain what was happening.

I should say in our defence that the same thing has been tried in England, Wales and Northern Ireland, but no one has managed to bottom out why there is such a variation.

It is interesting that the range is so similar across the United Kingdom—at the lower end, the difference is just a few pence. As John Simmons said, we cannot give the committee a definitive answer. I am sorry.

I seek clarification from John Simmons in relation to George Lyon's question. In essence, are you saying that wastage has been identified in relation to unreserved, rather than reserved meals?

At its simplest, the problem is that the ward orders more meals than there are patients to eat them.

To some extent, my question follows on from Margaret Jamieson's point about the so-called hidden subsidy for staff and patients. I do not want to be disrespectful, but I am mindful of the phrase about knowing the price of everything and the value of nothing. There seem to be many examples of such subsidies in the private sector—on oil rigs, for example. I have yet to meet an oil baron who objects to the subsidy of high-quality food that is sold at low prices often to a small minority of extremely highly paid people.

The situation is not transparent—I accept the logic of that in relation to attempts to establish whether there are other factors. If we are saying that there are hidden subsidies, does that mean that there are also hidden costs, for example in hospitals where there are no such subsidies? I am conscious that hospitals are often in isolated locations; it is virtually impossible for someone to go out and buy a sandwich at 3 am during a night shift at Perth royal infirmary or the new Edinburgh royal infirmary, for example. Is there a downside to going to the other extreme, as seems to have happened at the new Edinburgh royal infirmary, where people pay market prices for food? Was any investigation carried out to quantify what happens if prices are left to market forces, as opposed to there being an attempt perhaps to use a lower common denominator to help people with limited funds who visit patients—not to mention low-paid staff—to have a cup of coffee and a nutritious meal?

When we carried out the study, a number of catering managers made that point. The location of the hospital can mean that it is difficult for staff to go off site during a short lunch break, so staff might regard subsidised meals as part of their terms and conditions. We are more concerned about how hospitals manage their budgets if they do not know that they are subsidising meals. I suspect that some staff do not realise that they are getting a subsidised meal, because the situation is not presented to them in that way.

It is fair to say that we might have expected private contractors to rely on market forces, but we were unable to get costs from those contractors for the study. The costs that we have are from in-house providers only.

I want to go back to the point that George Lyon raised about the variation in the cost of meals. Was a range of costs apparent even within health board areas? Did issues arise from the procurement process? At the moment, each hospital has its own procurement process but, as we move towards having one system for each health board area, the situation might tighten up. Is cost linked to quality? The two trusts in Ayrshire and Arran have been highly commended and have received awards for the quality of their catering.

Within health board areas, there was no obvious link between the costs in different hospitals—although, if hospitals are near each other, they might charge the same. The majority of hospitals still follow the policy in the old national health service circular. However, some may have considered market forces. For example, if a hospital has a Tesco right next door, staff might go there. The hospital might compare its prices with the prices at Tesco, but it would not necessarily compare its prices with the prices at a hospital 20 miles away, because its staff could not get there in their break times.

I do not have the figures with me, but more than 90 per cent of the produce that is used by in-house providers in the NHS is procured through national contracts. That is done by the Scottish health care supplies division of the Common Services Agency. With national contracts, everyone is charged the same, although there may be local or regional variations. For example, hospitals in Aberdeen may want something slightly different, so they would negotiate with a local supplier. However, they will negotiate good deals, so the costs of such supplies would not really be an issue.

What about quality?

We have found no relationship at all between quality and cost. As Barbara Hurst said, we have tried to find every possible relationship, because there must be some relationships out there. However, we found no relationship between quality and cost.

I feel a very detailed study coming on in this area.

It is frustrating, because one desperately wants to find some sort of relationship.

I have always assumed that the high-quality food that is available in Ayrshire and Arran—which Susan Deacon will remember having tasted on a visit—costs more, but if the costs are within the range, that is brilliant.

We expected to find exactly what you suggest—high quality meaning high cost—but we did not.

In care homes, we found exactly the same thing. There were some high-cost, very high-quality meals, but there were also some medium-cost, high-quality meals. That pattern seems to be common.

You say that 90 per cent of procurement is done through national contracts. How is it possible to measure best value? Is it possible to compare supplies from national contracts with supplies from local suppliers? Can you use local suppliers within national contracts?

I think that it would be possible to use local suppliers through national contracts if that was what had been negotiated. John Simmons has told me off about this: I thought that it would be interesting to consider using local suppliers for catering, but John put me firmly in my place and said that we could not do that. He is probably right—it would be a policy issue for the local trust to decide whether it wanted to use local suppliers. Bigger issues arise to do with the local economy.

Can I tease that out a little bit? When decisions are being made about which suppliers to use, how easy is it to get a view of the best-value issues that surround local suppliers? Is it difficult to get information about the value of using local suppliers?

Do you mean whether it is difficult for a trust to do that?

Yes.

A trust would have a national contract because it would expect a discount for buying in bulk. It would therefore expect a local supplier to struggle to compete on price with that bulk buying, otherwise there would be no need for a national contract in the first place. As Angela Cullen said, a trust goes off a national contract because there is something in its local area that it wants to put on the menu to give it a local feel.

Is there nothing to stop a trust going off the national contract as long as it can use locally produced food within the framework of best value?

No, there is nothing to stop that; a trust could buy everything off contract if it chose to do so, but it would have to justify that by proving that best value was being obtained. As long as the trust could prove that the deal was as good as, or better than, the central contract, it would be fine.

Yes. It is difficult to define best value and I do not know what definition of quality is used. I understand that there are clinically based nutritional standards, but do they include whether a lettuce is crisp or whether fruit is fresh?

I say that a trust has to prove that it is getting best value, but it does not have to prove that daily to an outside body. Rather, it must satisfy itself that it is providing good-quality food to patients, so that the lettuce is as crisp as it would have been if it had been bought through the national contract, for example and comes at as good a price. The trusts have to consider the combination of quality and price.

On choice for patients, you mentioned that you considered the availability of alternatives such as vegetarian food or food for people from ethnic minority groups who have specific needs. Will you comment on that?

I was a bit surprised to discover that very few trusts complied with the Scottish diet action plan. Did you expect more compliance or that more action would have been taken following publication of the Scottish diet action plan?

I will let Angela Cullen answer the question about menu choice; she is an expert on menus.

We considered patient choice. The clinical resource and audit group—CRAG—guidelines of a couple of years ago said that there should be a three-week menu cycle and that choice should be varied enough that patients do not get fed up with eating the same food all the time.

When we considered patient choice, we found that there was a minimum of two main meal choices in every hospital that we examined in Scotland. That did not include salads and sandwiches because we counted hot meals as main meals. Obviously, if someone wants a sandwich because that is what they are used to eating at lunchtime, that choice is available. Therefore, if all the choices had been counted, there was a minimum of four: salads, sandwiches, and two hot meals.

We also considered options for vegetarians, special diets for people who have difficulty swallowing and eating, or who have special dietary requirements because of illness, and we considered ethnic minority meals. We found that there was a more limited choice of such meals. Some menus were very good and were dietary coded—which allows patients to choose the meals that are suitable for them—but other menus were not so good. There were suitable meals for diabetics, for example, but people might have had to ask the nurse which meals were suitable for them.

Although we found that provision of vegetarian options was a bit better, we also found in some hospitals that there was not a vegetarian option on the menu every day. A vegetarian would have to ask for a special meal to be provided. We found that in some circumstances the choice was very limited for vegetarian options because there was a weekly menu cycle rather than a three-weekly menu cycle. That is okay in an acute trust where the stays are short, but it is not good in a long-stay ward or hospital.

We also asked about arrangements for minority ethnic meals. We found that every trust and every hospital within each trust provided them, although some ward staff were not aware of the arrangements. When patients came on to a ward whose staff were not aware of the arrangements, they were not telling patients about them: the patients were choosing from the normal menu—as it could be called—rather than being told that there was another menu from which they could choose. There are issues.

It is a good-quality service. There is a decent amount of patient choice and a lot of hard work has been done, but communication is vital. Everyone needs to know that choice exists. If they want things that are not on the menu they have to know how to order them. There have to be protocols so that patients know to what they are entitled.

To follow on from Rhona Brankin's questions, I seek clarification. I think that you are saying that if a trust decided that it wanted food of a particular quality, that was organic or sourced locally or within certain degrees of freshness, nothing could prevent it from doing that as long as it could provide justification on environmental, nutritional or health grounds.

There is absolutely nothing that could prevent that.

At least one hospital—Edinburgh royal infirmary—is buying frozen meals, heating them up and serving them. From the point of view of your audit, were you able to form an opinion as to whether that option is likely to result in more or less waste, or is there no difference?

There is no difference, because we were considering waste where food was coming to the ward. How food is delivered to and how it is ordered from the ward are what make the difference.

Can I clarify something? Robin Harper said that there would be nothing to stop a trust buying local freshly produced food if it wanted to, but that is only within the current definition of quality, as contained in your best-value definition.

Can you run that past me again?

You say that a trust can buy whatever quality food it likes, so long as it is within the best-value framework. I am trying to tease out the concept of best value, and the definition of quality that you use when it comes to food, within the definition of best value, because therein lies the nub of the issue. One person's definition of quality of food might not be the same as somebody else's. That is why I was trying to tease out the whole business of nutritional value. A cabbage that is two weeks old might have similar nutritional value to one that is a week old, but there might be a hell of a lot of difference in taste. I am particularly interested in that, and in the actual flexibility that trusts have.

If I go into hospital, I would like to have the option of eating high-quality locally produced food whenever I could. That is part of my definition of quality. I might get a frozen meal from Wales or wherever, and nutritionally there might not be much difference, but—

I think that I now understand your point. Best value is not simply about the cheapest way of getting a product; there has to be a balance between cost and quality.

In this study, we asked patients about quality, because they are the best arbiters of what they are getting. We also examined nutritional content. Although the question of locally sourced and produced food could well come under the umbrella of best value, the issue is to do more with local planning than simply with the meal on the plate. If a trust was involved with local businesses in developing local food of the right quality, such an approach would come under the definition of best value. However, we did not consider that issue. The assessment of quality rested purely on patient satisfaction and nutritional content. That said, I agree with your comment that a wider definition of best value might include other factors.

I must apologise for my arriving late this morning.

Apology accepted.

If I ask about an area that has already been covered, convener, I am sure that you will tell me so and I will move quickly on.

The report identifies different areas where waste might occur such as unserved or uneaten meals. Although a lot of attention has been paid to factors that contribute to meals' being wasted because they are not served to patients, less attention has been paid to the issue of wastage because of food that has not been eaten. Will you comment a little more on that matter?

Moreover, will you comment on the question of assistance with eating? We know from a range of studies in other areas—not least in relation to the nutritional status of elderly people—and from all sorts of anecdotal evidence, that one of the biggest problems is that although food is being served to frail and elderly people, they do not receive physical support to eat it, so it remains uneaten. In a sense, the quality of the food on the plate does not matter if it is not going into the patient's mouth.

The issue of uneaten meals is quite interesting. As far as wastage is concerned, I must be honest and say that we felt that we could not ask our auditors to assess what was left on people's plates.

As you said, the problem is either that people need help with eating the food, or that they cannot stand what they are given anyway. We tried to examine the latter option through patient satisfaction. However, we left the former point because NHS Quality Improvement Scotland is developing standards that should cover assistance with eating and the question whether people are being given the right help, the right implements and so on. That said, a local report from England has proposed closing wards for a couple of hours over meal times to allow people to receive help with eating. The issue is important, but it is outside the scope of the study.

My next question is on the separate area of guidelines, standards, specifications and so on. In your report, you recommend that

"National catering and nutrition specifications should be developed for NHSScotland."

However, one of your key findings is that

"around three in five catering specifications do not fully comply with the model nutritional guidelines".

In this and in other areas, I am slightly anxious that the response to gaps between knowledge of what is good practice and actual practice is simply to develop yet more standards instead of to focus on good performance against existing ones. My concern is that if a lot of time, energy and effort go into developing yet more specifications—much work has already been done in that area—it could divert attention from improving practice. Can you give us any assurances on that front or tell us at what stage is the development of nutritional specifications? Perhaps we can get some assurance that that work will add value to existing practice.

You are absolutely right—it is a busy field. Not only is there our report; there are the NHS Quality Improvement Scotland standards and some work has been carried out by the Health Department around those specifications. Angela Cullen might want to say where the Health Department is at with that work.

One of the reasons why we made the recommendation is that we found that three in five hospitals were not complying with the model nutritional guidelines in the Scottish diet action plan that Rhona Brankin mentioned earlier, but which we did not address.

Although we did not undertake a full and detailed review of nutrition, we picked up on the feeling that the dietetics service is under-resourced. The reason why some places do not have a nutritional specification that complies with the model nutritional guidelines is that they do not have dieticians in place to ensure compliance. We felt that a national nutritional specification could be developed for local implementation. Some places have good dieticians, so it could just be a case of those places sharing what they have with other places.

We do not think that it would take a lot of work to develop a full national nutritional specification that sets out basic things such as how many calories should be consumed in a day, how many grams of protein should be consumed and what the menu and meals should contain. That could be rolled out across the service and local catering departments could examine their menus to ensure that they comply with it. We should have a national specification because it would cut down on work rather than create more.

Susan Deacon is quite right to say that there is no point in developing more standards unless they are picked up on. However, our local reports suggest that that is not happening. Those reports are followed up much more quickly than the national report will be and some pressure is being exerted through that line to ensure that trusts and hospitals comply with existing and new standards. We see that as very important.

The approach seems to be one of uncharacteristic caution or subtlety as a means of achieving change. Why do your recommendations not say that all trusts must comply with the model nutritional guidelines as stated in the national action plan that was published in 1996? Why do you not recommend that trusts should assess provision of dietician support and increase that support when necessary? Would that not be a more direct way of reaching the same point? I am playing devil's advocate.

You are definitely playing devil's advocate. You are absolutely right. We were trying to pick up the issue within the context of what is happening at the moment. The Health Department has a steering group that is working on nutrition. Angela Cullen may be able to give a bit more detail on that. We also have the NHS Quality Improvement Scotland standards. There is a lot of push at the moment.

We should not lose sight of the fact that many hospitals are doing good work on this. We did not want to over-egg it, but we wanted some sort of national push. To tell you the truth, I think that once the national report comes to the committee, it has the stamp that allows it to drive things quite quickly. Angela Cullen can pick up on what the department's group is doing.

The departmental implementation group has been in existence for the same period as we have been undertaking the study and NHS Quality Improvement Scotland has been developing its standards. QIS aims to ensure implementation of its standards, but another issue that it has been considering is developing a national nutritional specification. It has started work on that, but it has not yet got to the stage at which it is ready to develop something. It has pulled together some people who have already done some work on standards that can be developed nationally. As a specification is already being developed that can be published for the whole of Scotland, we are not asking for something enormous to be done. The ball is already rolling.

I am genuinely grateful for those answers, but I still think that we are heavy on the process and a bit light on the results.

I believe that a recommendation is missing on the measurement of progress. There is no recommendation that the Executive should include nutritional standards and so on in the performance assessment framework—the tick box that I keep on talking about. Why have you not made such a recommendation, so that there would be departmental monitoring of progress? When do you envisage doing a follow-up report to the baseline report?

The PAF issue is an interesting one, because every time that we do a report there is a huge temptation to say that the situation should be reported through the PAF. We have had dialogue with the Health Department on how many indicators it has on the PAF. There are a lot of indicators, so the department's understandable position is that if more are added, some will need to be taken out. We are cautious, as you have identified, about throwing all our reports into that mechanism, but you are right that we could have done that.

On when we will produce a follow-up report, we think that 18 months might be a decent time interval to allow for things to have happened on the back of the report.

Thank you for that comprehensive session on "Catering for patients", which I do not think that we over-egged at all. [Laughter.] That is what you call a "Week in Politics" moment—just like crisp lettuce—but "The Week in Politics" is not on this week, so we will not see it.