Management at the London Ambulance Service made ‘every mistake in the
book’ in procuring, installing and operating its £1.5 million computer-aided
dispatch system, says the report of the four-month inquiry into why the
system collapsed last autumn.

It adds that managers were under enormous pressure to get the system
up and running within the ‘unrealistic’ 14-month deadline they set themselves
in an attempt to improve emergency response times. This led to the system
being introduced before it was fully tested and before staff had been given
adequate training.

The health reforms which have turned ambulance services into ‘service
providers’, bidding for work from hospitals, could cause a repeat of the
errors that led to the collapse of the LAS dispatch system.

The LAS is the largest ambulance service in the world, serving a population
of 6.8 million. The pressures it faces may also affect other ambulance services.
Because NHS reforms force them to bid to carry patients to and from hospitals
– 95 per cent of their work is routine transport rather than emergencies
– some regional services have decided that it is cost-effective to merge,
or in some cases to try to ‘poach’ work from adjoining areas.

The LAS received 900 complaints about its service on 26 and 27 October,
when it switched over from semi-computerised methods of allocating and tracking
its ambulances, with paper print-outs confirming data on screen, to an entirely
computer-driven system, without paper records. From 7 am, when the system
was put into operation, it began ‘losing’ ambulances because information
coming in via radio links with ambulances was incomplete. As a result, the
system began allocating more than one ambulance to an incident.

The problems snowballed, but without paper records control room staff
could not identify the system’s errors. Six hours after the changeover,
the volume of calls to the control room was rising fast and it was taking
10 minutes to answer them. But early claims by the LAS that these were busy
days were incorrect: the number of patients carried show it was an average
day. The excess calls came from people who became worried when help did
not arrive.

On these two days, fewer than 20 per cent of ambulances reached their
destination within 15 minutes of the call being taken. The average for October
was just over 30 per cent, already a very low figure. In May, before introducing
the semi-computerised procedures, the LAS was reaching more than 65 per
cent of incidents within 15 minutes. Government guidelines say the figure
should be 95 per cent.

After 27 October the LAS went back to its semi-computerised methods.
But shortly after 2 am on 4 November the system crashed because a programming
error had been allocating nonexistent tasks to the computer’s memory for
three weeks. The LAS reverted to manual methods. Figures released by the
trade union NUPE show that in January about 64 per cent of ambulances reached
their destination within 15 minutes.

The inquiry said that examination of coroners’ reports going back to
November 1991 found that none of the 26 deaths in cases where the LAS was
called to give evidence could be attributed to the late arrival of an ambulance.
Two cases are still outstanding.

Paul Williams, a member of the inquiry team, says that ‘the majority
of the hardware’ from the failed system could be retrieved for future use.
But he estimates it would take five years to introduce the system the management
had envisaged, to allow for proper testing and staff training.

Mergers have reduced the number of amublance services from 62 in 1990
to 60. Recently the Liverpool and Cheshire services merged to form a single
operator covering 2.3 million people. The services in Bedfordshire and Hertfordshire
have done the same. Others are known to be touting for business in adjoining
regions. ‘It’s for economy of scale,’ says Paul Goss, of McDonnell Douglas
Health Systems, which supplies computer-aided dispatch systems to 18 services.

The two mergers have gone smoothly, but merging always carries the risk
of systems failing during the changeover. If the original systems are incompatible
a new one will be required, and staff need to be trained in unfamiliar methods.
If the systems are broadly compatible, small differences can confuse staff
at critical periods. Attempts to ‘bolt together’ the existing systems rather
than buying a new one could create fresh incompatibilities.

Clinton’s decision to send the space station back to the drawing board
did not surprise ESA officials. The station has been shrinking year by year
because of budget cuts. The consequent redesigns of Columbus have continually
forced up its cost and delayed its launch by two years.

ESA managers say this latest redesign promises to be the greatest upheaval
of all. ‘We are worried, no question about that,’ says Franco Emiliani,
programme manager for Columbus. ESA officials say they have received little
information about the scope of the changes. Emiliani says ESA will postpone
the final go-ahead for Columbus until the redesign is complete.

This latest redesign makes cancelling Columbus in favour of greater
cooperation with Russia more attractive. The Russian space agency has invited
ESA to provide a laboratory module for the Mir-2 space station, scheduled
for launch in 1996. ‘We would certainly look at that option,’ confirmed
Daria Robinson, an ESA spokesperson.

ESA has already shifted research on its spaceplane Hermes into a joint
project with the Russians.

In addition to its liquid-fuelled main engine, Ariane 5 will have two
boosters strapped to its sides which burn solid fuel. The firing of the
first 31-metre booster is scheduled to take place on a specially designed
test stand at ESA’s launch site in Kourou, French Guiana.

Ariane 5’s first flight was pushed back six months to October 1995 after
problems in solidifying the 237 tonnes of rubber-like propellant for the
booster. This and other difficulties have pushed the project 10 per cent
over budget.

Ariane 5 was conceived along with the space shuttle Hermes and the Columbus
space station programme during the mid-1980s. In the harsher economic climate
of the 1990s, European space ministers have shelved Hermes and part of the
Columbus programme.

Ariane was intended to lift Hermes into orbit, now its sole role will
be to launch satellites. The rocket can lift a single satellite weighing
up to 6.8 tonnes or two totalling 5.7 tonnes.

ESA believes Ariane 5 will cut the cost of launches by up to 20 per
cent by carrying two satellites at once. Many in the space community doubt
this will be possible. ‘It will be very, very difficult,’ said one European
space official. As Ariane 5 was designed to lift astronauts into space,
it has extra safety features that make it more expensive, he said.

In addition, satellites appear to be getting lighter. The telecommunications
satellites Intelsat 7 and 8 will be lighter than the 4-tonne Instelsat 6,
which was the heaviest satellite raised into orbit. The latest satellites
for Inmarsat, the marine satellite communications organisation, weigh less
than 2 tonnes.

ESA is convinced the Ariane 5 will find a market. And Jacques Durand,
the project manager for Ariane 5, says the extra safety features of Ariane
5 should achieve 98 per cent reliability, compared with the 90 per cent
reliability of Ariane 4. ‘It should give confidence to the insurance world.’