Shot down

In August the chief medical officer announced the extension of the influenza vaccination programme to all people over 65, with a minimum acceptable uptake in the first year of 60 per cent.

1Health authorities are being given£10m for the programme and have been asked to work with local GPs to achieve the target overall in their districts and to monitor the uptake in the 65 years and over age group.

1How ready are general practices for this, and how realistic is the target take-up?

In 1996, the most recent year for which published data exist, 23 per cent of people over 65 years old and 44 per cent of those at high risk - the other main group eligible for flu vaccination - had the vaccine.

2If these figures are to be improved, considerable improvements have to be made to the programme.

In 1997 the US coverage of those aged 65 and over was 63 per cent.

Health authorities have been allocated the extra funds to help identify those eligible, and to monitor the progress of the vaccination campaign, as well as to help practices with insufficient staff. GPs will also receive an item-of-service payment for vaccinations given, but for this year only.

Last year we studied nine general practices in England to look at their vaccination programmes.

All practices were in urban areas with between two and 14 partners, and practice populations ranging from 3,000 to 24,000.All had practice nurses and access to district nurses. All the practices were computerised, although five of them were still using some paper records. We contacted the practices through the Medical Research Council's general practice framework. We visited each practice and interviewed key staff, the nurse who organised the clinics, the practice manager, and sometimes the GP.

Identifying those eligible A rational approach might suggest that the first step would be to estimate the number of vaccinations required. This would require interrogation of an age-sex register, some method of identifying the 'at risk' groups, and the likely take-up rates. This information is crucial for ordering vaccines, communicating with those eligible for vaccination, and auditing the programme.

In the practices visited, representatives of pharmaceutical companies contact practices, usually in January, to take vaccine orders for the following autumn. Practices felt that it was difficult to vary orders subsequently. All practices in fact ordered on an incremental basis, based on the number of vaccines used previously. When asked if the guidance to vaccinate all those aged 75 years and older in the summer of 1998 had affected their orders for 1999, only two practices had ordered extra.

Communication Recruitment depends on having an information system able to list those eligible. Only three practices were able to identify all those in at-risk groups at any age. Although all practices had an age-sex register, none used them to plan their programmes.

The practices used mail-shots, notices in surgery and labels on repeat prescriptions to contact patients. Six of the practices relied on patients to present themselves for vaccination. One mailed all those eligible and the other two contacted non-attenders. But response rates among non-attenders were low. Patients often took the initiative by asking for appointments, and in one or two practices even signing a list in the summer, asking to be allocated an appointment on the vaccination clinics.

Seven practices did not audit their flu vaccination programmes, and while all were able to quote the number of vaccines given, it was not possible to ascertain the number by age and at-risk group.

Seven practices ran clinics with individual appointments and two had drop-in clinics when patients queued for up to an hour for vaccination.

Most practices used a computer system for appointments or logging those who had a vaccine.

Other practices used patients' notes or a copy of the appointments sheet to record vaccination, and often entered this on to the computer later.

Delivering the vaccinations There were three main methods of delivery.

We dubbed the first 'super Saturday' - a Saturday morning clinic organised by practices having an active individual approach to recruitment, recording and follow-up.

The second, 'chain store approach' consisted of half-day clinics. The third was the 'local store' approach, with the practice nurse allocating nurse time in the clinic for vaccinations.

Two practices did Saturday clinics. In one, all staff in the practice who could vaccinate were recruited to participate in a clinic held over two to three Saturdays. In the other practice, which initially recruited passively and then mailed nonattenders, a drop-in system was used.

Chain store - special session clinics In this model, regular clinic sessions run by practice nurses were used as flu clinics, on specific days, for six or more weeks in October and November. These clinics tended to displace routine work, and temporary space was made available. Two people ran the clinics.

One called in patients, checked eligibility and logged in the vaccination. The other checked for contra-indications, asked them to bare their upper arms, gave the injection, and prepared for the next patient. One of the clinics organised in this way had split the tasks into two rooms.

One was used by an assistant to prepare up to four patients for vaccination and record the vaccination, with the nurse then vaccinating them and seeing them out, as four new patients entered the other room. If an eligible or at-risk patient was identified, they would be vaccinated there and then by a GP.

Each received reimbursement and dispensing fees under the personal administration fee system.

Practices also tried to obtain bulk purchase discounts from vaccine manufacturers. This discount, with the reimbursement and dispensing fees, formed the direct recompense to practices participating in the programme. Arguably practices also benefited considerably by reducing the workload during the flu season.

Costs and efficient practices We collected information about: the time costs of staff for ordering, planning, communicating with patients, vaccinating patients and recording work done; the overhead costs of mailing patients and maintaining the premises used when vaccinating people; the capital costs of maintaining a cold chain refrigeration system for a large increase in the number of vaccines in a short space of time, and of consulting room space use.

There were considerable opportunity costs to undertaking the programme, which we have attempted to estimate. The capital costs associated with the programmes were estimated and discounted over the likely useful life of the asset.

Cold chain systems, for example, had to be in place in several consulting rooms either far apart or in separate premises and had to be replicated.

The ordering of drugs was an important aspect. If the order was not accurate, extra costs would be incurred. The negotiation of bulk discounts was a commercial issue and so individual practice costs were not collected, but a common cost per dose used (£2.60), illustrates the fact that at least 50 per cent of the cost of flu vaccination was the cost of the vaccine in all settings.

The various models of delivery of vaccination had different throughput times. The model used by one practice of vaccinating four people at once administered more vaccinations per fiveminute intervals, but was more costly because of office space used on a weekday.

A clear cost-effective method was not identified. But some indications of cost areas to be considered could be identified. Staff costs for vaccine delivery were among the highest costs. The practice actively recruiting patients and running Saturday clinics in a short period of time had one of the highest staff costs per vaccine but also had the highest doctor/nurse ratio. The highest costs for outreach were in those practices with the highest proportion of special visits just for flu vaccination.

The process is labour and vaccine intensive.

Active recruitment was the major factor in the variation of the overhead costs, but overheads, with this activity included, were at the most only 50p per vaccine delivered. Capital costs were small.

Variations were mostly due to refrigeration not being optimally used - for example, in a small practice, or where batch deliveries of vaccine from manufacturers were not feasible.

As suggested, costs per vaccine given appear to vary with a number of factors. A firm assessment of the key ones could not be determined from the small numbers in the case study. The systems using paper records during the vaccination process seemed the more expensive ones. Extra administrative time is needed to get notes out, call patients in from reception and process the data.

Lessons The research suggests a number of things that could be done to improve the delivery of the flu vaccination programme. The new Department of Health programme suggests health authorities may take on the role of directly informing eligible patients that a service is available. An active, individualised approach is being considered, similar to that already adopted by some practices.

The well-organised practices which already communicate directly with patients may not find this useful: it may limit their ability to book patients for their special clinics and they may be concerned in general if a centralised method was seen to be impersonal or of poor quality.

If the health authority initiative were applied only to practices that would 'struggle to achieve uptakes', inequalities among practices would be introduced.

The proposed use of local and national media, with posters and leaflets, would benefit all practices.

The influenza vaccination programme has administrative difficulties in identifying and targeting people at high risk. Computer-assisted generation of lists of eligible people, and the use of reminder cards, increased the coverage in Trent region but were implemented infrequently in 1992.

Ofthe nine practices we visited, only one out of the four with no manual records actively contacted all eligible patients. Computerisation as such, therefore, seemed insufficient to mobilise the practice to undertake active recruitment. The policy proposals to mandate HAs to monitor the vaccination programme should change this, by strengthening the programme and enabling weaknesses to be identified. Perhaps this is the most useful of the new policy proposals.

Our sample also indicated staff availability as a critical point. More resources for district nurse sessions for visits to residential care homes, and additional practice nurse sessions to help 'struggling GP practices', are being proposed.

Outside general practices, some high street chemists are offering the vaccine to those eligible outside the targeted programme. This may be more convenient for active people in the eligible group.

A one-off item, service payments, is being brought in as an incentive for vaccinating 65-year-olds and older this winter, but not for those at risk at other ages. At£6.45 the fee is higher than the highest estimated from our sample of nine practices which includes vaccine costs that are reimbursed. Thus practices already able to run large flu vaccination clinics will benefit most. Incentives to vaccinate those at risk in younger groups might prompt practices to identify and communicate more effectively with those at risk.

Although administrative difficulties are the main factors behind low flu vaccination coverage, professional or public attitudes may also be behind some of them.

In those at high risk, a lack of acceptance of influenza vaccination explains little of the perceived variation. In a study in Leicester, 80 per cent took up vaccination when it was offered during visits to their doctor for another reason.

6The research, undertaken in 1992, found that less than 30 per cent believed that vaccine reduced mortality or complications a great deal.

6Our research suggests the cost of actively recruiting patients may not be a large part of the overall programme costs. The extra£10m to improve flu vaccine coverage, this year only, can also be used to provide extra staff, but the sustainability of this key aspect has yet to be faced.One-off funds to monitor uptake will help practices and HAs to develop good surveillance of vaccine coverage.

REFERENCES

1 Chief Medical Officer. Influenza Immunisations. 1 August 2000. Department of Health PL/CMO/2000/3.

2 Irish C, Alli M, Gilham C, Joseph C, Watson J. Influenza Vaccine Uptake and Distribution in England and Wales, July 1989-June 1997. Health Trends 1998; 30: 51-5.

3 Wiselka M. Influenza: diagnosis, management and prophylaxis. BMJ 1994; 308: 1341-5.

4 Nicholson KG, Wiselka MJ, May A. Influenza Vaccination of the Elderly: perceptions and policies of general practitioners and the outcome of the 1985-86 immunisation in Trent, UK. Vaccine 1987; 5: 302-6.

5 Nguyen-Van-Tam J, Nicholson KG. Influenza Immunisation: policies and practices of general practitioners in England, 1991/1992. Health Trends 1993; 25: 101-5.

6 Nguyen-Van-Tam JS, Nicholson KG. Influenza Immunisation: vaccine offer, request and uptake in high risk patients during the 1991/92 season. Epidemiol.Infect. 1993; 111: 347-55.