After Claudia Comberti's inquest

On Thursday 26th October I attended the inquest of Claudia Comberti, a DPhil student at the University of Oxford. She had died on the 9th May 2017 while cycling on the Botley Road after falling from her bicycle into the path of a double-decker bus operated by Oxford Bus Company.

I found the implications of the inquest somewhat problematic, and I'll present these concerns in the context of what happened at the inquest. If you're short on time, you might want to skip straight to the conclusion.


To be clear, this blog post reflects my personal recollection of the inquest and my own thoughts around it, and does not necessarily reflect the views of any group or organisation with which I am associated.

The incident

Claudia was heading back into town from a WildCRU lecture, cycling eastbound along the Botley Road through the junction with the A420 link to the A34.

She approached the junction in the left-most shared bus and cycle lane while the light was red, and passed the line to wait a couple of metres beyond for the light to change, about a ⅓ of the lane width from the left.

The bus behind slowed for the red light, but the light changed as the bus approached and so the driver did not stop. The bus's recording equipment showed that the bus's speed dropped to 7mph before the driver started accelerating again.

CCTV still of Claudia Comberti from the following bus

The first of two stills from the bus's CCTV, showing Claudia just after the light had changed.

Claudia set off as the light changed, and moved further right towards the implicit centreline of the left-most lane across the junction. The conclusion was that she'd been attempting to accelerate from the lights, and that her right foot had slipped off her pedal, throwing her rightwards. Her bike had clipless pedals, and she was wearing normal shoes, which may have meant less grip between her shoe and the pedal. Her chain had come off, but the forensic investigator couldn't be sure as to whether this had happened before or after the slip.

The bus driver behind had decided to overtake Claudia on the right within the bounds of the junction, but Claudia's foot slipped when the bus was about 6-8 metres behind and directly in front of the bus. The driver attempted to move the bus further right to avoid her, but she fell an estimated 2.3m to the right and was hit by the nearside front of the bus — cracking the body panel — before going under both the front and rear nearside wheels. At the time of impact the bus was travelling at 15mph, and the driver did not start to apply the brakes until after the collision.

The forensic collision investigator determined that the bus driver had about a second in which to react, that there would not have been enough time to avoid the collision by braking, and that attempting to steer away was a reasonable response.

The Oxford Bus Company CCTV operated at a frame every 0.48 seconds, which seemed to make it more difficult than you'd hope to be sure how the incident unfolded. I'd hope they could be pressured to improve this to make collision and incident investigation easier.

Evidence at the inquest


It's worth remembering at this point that this was an inquest, to determine the circumstances under which someone has died, and not to determine criminal or civil liability. By this stage it's already been determined by the police that no one's driving falls below that expected of a competent driver (i.e. the lowest threshold for a careless driving charge). I may be wrong, but I assume this is the reason that the coroner doesn't particularly challenge the driver of the bus.

The coroner read evidence from a number of witnesses, with two witnesses — the driver of the bus, and a police forensic collision investigator — giving evidence in person. Instead of considering each witness in turn, let's pick out the key themes in the evidence.

Pressure to make progress?

It became apparent that bus drivers know the junction to provide a good opportunity to overtake cyclists. The road layout forces cyclists off of a shared-use path into the carriageway bus lane on the approach, and there's then no cycling infrastructure that a cyclist can use to get out of the bus lane on the other side. If a driver can't overtake within the junction then they can end up behind the cyclist for a period before another opportunity to pass becomes available.


Cyclists are forced into the bus lane to traverse the junction


After the junction, cyclists are forced to remain in the bus lane, which slows bus progress if they are unable to overtake.

The driver of the bus agreed that the junction was an opportunity to overtake, saying that another bus behind is a factor, not necessarily so that they can also overtake, but to give that driver an opportunity to see the cyclist and so know not to follow through blindly. He also said "you do overtake cyclists to get to the next bus stop".

The police forensic collision officer said that the bus driver could have held further back, but that he could understand why he would try to overtake at the point he did, and "competent drivers would choose to overtake".

A desire to make progress shouldn't be a factor in deciding whether or not it is reasonable to attempt a particular overtaking manoeuvre. And yet, the police called it "understandable" and something a "competent driver" would consider reasonable.

Of course, this desire doesn't come out of nowhere. The bus driver is running to a timetable and needs to "get to the next bus stop", as the driver at inquest said. Tom Kearney, a London-based bus safety campaigner, writes in 2014 of Transport for London's Excess Waiting Time performance metric:

I am convinced that TfL’s performance measure of Excess Waiting Time (EWT) is a particularly dangerous feature of TfL’s performance-based Quality Incentive Contracts. EWT is a number that is calculated by the average gap in-between buses over a contract-determined performance period (this is called a Quality Service Indicator “QSI” period). An EWT indication of “0” reflects perfect in-between bus spacing (remember, the time gap between buses is the important part) and, on this basis, BusCos are given QSI targets to meet by TfL. A Higher EWT number means the Bus Driver has room to manoeuvre on a route for them to meet the QSI performance target.

While TfL have focused on EWT as neutral indicator by which it can objectively measure performance, I believe this contractual focus is to blame for the "environment of urgency” forced on Bus Drivers every day. By focusing on EWT on the hierarchy of system needs, TfL appears to be ranking performance higher than safety. I fear that this foolish devaluation of safe driving has cascaded all the way down from Management and can influence how Bus Drivers behave on the road. While I’m informed that many Managers and Controllers consider TfL exercising a three month break clause in a bus contract as the ultimate punishment for poor performance, I’ve never heard the same angst expressed by them about a rash of bus collisions resulting in pedestrian or cyclist KSIs along the particular route. Priorities. Priorities.

While I have no knowledge of how Oxford's bus companies approach performance targets, I would be very surprised if they didn't have their own "environment of urgency" to some degree that compromises safety. My own experience in Oxford is of some drivers feeling unable to keep a safe distance or wait for a safe opportunity to overtake, often passing or following too close.

Overtaking with enough space

The coroner asked the driver, "What is training with regards to overtaking cyclists?" "It's probably covered in hazard perception; you use common sense." No mention of the Highway Code (Rule 163 — "give motorcyclists, cyclists and horse riders at least as much room as you would when overtaking a car"). No mention of Thames Valley Police's recommendation less than a month earlier to allow at least 1.5m.

The coroner followed up by asking about minimum overtaking distances: "I'm not aware of one, but [you should] give as wide a berth as possible." But what does this actually mean? Would "I needed to get past, but I could only give them a foot, so that's what I did" be acceptable?

It's staggering that in a city full of cyclists that a bus driver cannot say that they've been trained or received specific guidance about overtaking them safely. This needs addressing with the bus companies and the city and county councils.

Furthermore, if Thames Valley Police actually took enforcement action against drivers who overtook too close, then maybe drivers would become more aware of how close they were passing vulnerable road users. West Midlands Police have reported a 20% reduction in the number of cyclists killed or seriously injured in the county since running their Operation Close Pass education and enforcement operation.


It was implicitly acknowledged at the inquest that the junction and surrounding road design creates conflict between cyclists and buses, but no blame was attributed in this regard.

The traffic management officer who had attended the scene had provided written evidence to say there were no infrastructural contributory factors, but I assume this refers to e.g. surface issues, and not the wider design. He also said that there had been three other incidents involving cyclists in the past three years, but this doesn't tell us much. Do cyclists avoid the junction because it's perceived as dangerous? How frequent are near misses? How many other times would overtakes have resulted in collisions if a cyclist had slipped or had a wobble at just the wrong moment — the very thing the Highway Code provides as a reason for giving cyclists plenty of room?

As for the other incidents, there is one in which (I assume) a car driver failed to give way when turning into the access lane for the park and ride, and I can't work out what the others are. However, there are a number of cases of cyclists being hit on the shared-use path by vehicles entering and leaving the main road east of the junction.

More generally, designing out conflict is the only practicable way to ensure that slips, misjudgements and "momentary lapses of concentration" do not result in deaths and serious injury. If cyclists had a segregated route across this junction and a protected cycleway along the Botley Road then we create an environment that is, and feels, safer for cyclists, and eliminates the pressure for bus drivers and others to risk unsafe overtakes.

The coroner had written to Oxfordshire County Council to ask whether the junction could be improved, and they had put in a funding bid to improve the cycling infrastructure on the Botley Road. It had been announced the day before that the council had secured £5m from the DfT:

About £6.8m will be spent on improving how pedestrians, cyclists and buses use Botley Road between Binsey Lane and Eynsham Road as part of the council’s Oxford Transport Strategy.

Cyclists will be segregated from the main road and pedestrians on most of that route.

This is promising, and I'm sure Cyclox and other interest groups will keep a close eye on the proposals as they take shape.

It was on this basis — that the County Council were seeking funding to improve the infrastructure — that the coroner did not feel it necessary to issue a Prevention of Future Deaths report. I feel that this misses an opportunity to further investigate or improve the other issues (driver training and performance pressures, and overtaking safety).


There's a lot here, but if you're after take-away messages — and you should be — then:

  • It shouldn't be acceptable, societally or in the eyes of the police, for drivers to put themselves in situations where they don't have enough time or space to react to a cyclist coming off. If you've already given 1.5m, an extra 80cm is much easier.
  • Someone needs to look at how bus companies in Oxford train their drivers to behave around cyclists. For a bus driver to be so vague about what training they've received and safe passing distances is deeply worrying.
  • Oxford Bus Company needs higher frame-rates — i.e. better than 2.08fps — on their bus CCTV to better assist investigations.
  • Thames Valley Police need to be taking enforcement action against those who risk the safety of vulnerable road users, even if there's no collision or injury. West Midlands have pioneered this, and have a 20% reduction in cyclist KSIs. The enforcement creates a change in behaviour far beyond those that are prosecuted, and as a result there are people still living and walking now who wouldn't otherwise have been.
  • Infrastructure. Oxford needs continuous, segregated and protected cycle routes along key routes to design out the kinds of conflict that leads to incidents like this one.
  • A slip, a pot hole, a not-getting-unclipped-in-time, a misjudged overtake should not result in a death or serious injury. It should be designed for.

News coverage

Other responses


A scanned copy of the notes I took during the inquest hearing are available here. If there's any doubt as to the interpretation of them, please contact me to check.