Forum Topic

LTNs: new peer-reviewed articles on effect on emergency services and crime

These two articles puncture a couple of myths:https://findingspress.org/article/23568-the-impact-of-2020-low-traffic-neighbourhoods-on-fire-service-emergency-response-times-in-london-uk"These findings add to evidence that LTNs do not adversely affect emergency response times."https://findingspress.org/article/23623-short-term-association-between-the-introduction-of-2020-low-traffic-neighbourhoods-and-street-crime-in-london-uk"Overall crime trends in and around LTNs were more favourable than the background trend in Outer London, and similar to or slightly more favourable than the trend in Inner London. This pattern was also seen for numbers of direct attacks against the person - and this may underestimate the benefit per pedestrian, given evidence that LTN introduction is associated with increased walking."

Tom Pike ● 1993d34 Comments

Looking at the abstract the report seems to work on the assumption that the reduction in traffic was due to the LTNs. Isn't it more likely that there were fewer delays to emergency services due to traffic because traffic was down during the lockdown?

Francis Rowe ● 1993d

I think traffic was actually at pre-pandemic levels quite quickly - by mid-year - while public transport journeys were down - perhaps because people were shunning public transport and driving instead.

Ed Saper ● 1993d

The argument that's surprised me is that roads with lots of vehicle traffic are assumed to be safer than roads without.I could understand if it was completely deserted, but if more people are walking and cycling then I would have thought it would make roads safer.Inside the car you're not necessarily going to notice someone (particularly at dusk / night) - having more people on foot or relatively slow bikes who can see and hear better I'd have thought would act as a better deterrent than a passing car.

Ed Saper ● 1993d

From the abstract it does not appear as though any consideration of traffic volumes formed part of the study.  It makes the study essentially meaningless.Yet despite this major lacuna in the study it stated "there was an increase in the proportion of incidents coded as ‘delayed due to traffic calming’ (from 1.2% to 3.4% inside LTN areas, from 1.6% to 2.1% in the rest of London)"My problem with that statement, interesting as it is, is that there is no indication of whether this increase is statistically significant or just natural variability.The study then goes on to state "This was, however, more than offset by a decrease in the proportion of incidents coded as delayed for other reasons, particularly ‘delayed by traffic, roadworks etc" - this is just another example of how the study just does not seem to be looking at the factors that may or may not affect response time correctly.In fairness to the authors of the study I would emphasise that I read the abstract not the study and it may be that the study addresses what would otherwise be a completely flawed analysis

Justin Stephenson ● 1993d

You should read the full article, as have the expert reviewers that assessed this work for any obvious flaws might be identified by just reading the abstract.This research is looking at changes in emergency response times, not their absolute values which of course will depend on the level of traffic. It concludes that there is no statistical difference in the changes within LTNs, around LTNs or elsewhere.

Tom Pike ● 1993d

The concern was that response times had rocketed during the last year due to LTNs delaying emergency vehicles, despite traffic volumes thought to be low:https://www.dailymail.co.uk/news/article-9413629/Traffic-calming-measures-trigger-spike-fire-service-delays.htmlThis is based on anecdotal evidence and the increase in "delayed due to calming" code use.However, overall delays were slightly down during that period.That's important because the frequent criticism is "if emergency vehicles are delayed now, think about what it will be like when the traffic goes back to normal".But they weren't delayed - they're slightly quicker.

Ed Saper ● 1993d

As instructed I've read the full report and the abstract was indeed an accurate reflection of the content. The assumption appears to be that traffic fell in LTNs in 2020 due to their implementation and not lockdown.Nearly everyone is going to accept that if emergency service workers say they are being delayed by LTNs then emergency service workers are being delayed by LTNs. The Jesuitical arguments employed by the authors of this report that the evidence of the emergency service workers is anecdotal and misrepresents what is happening is going to be discounted by everyone except those who don't want LTNs to be criticised in anyway.By trying to cast doubt on something that most people would regard as self evident, it seems to me that proponents of these schemes are picking the wrong fight. The first question that should be asked is not 'do LTNs delay emergency vehicles?', they clearly do, but why are councils allowed to make material changes to road layouts without providing the necessary data to the emergency services so they can alter their routes? This would immediately make LTNs neutral in terms of their impact on emergency service journey times.For the record I do think LTNs will become increasingly essential, particularly in this part of London where we are going to see a significant rise in population density but that technology should be used to mitigate harm.In Chiswick specifically don't think LTNs have significantly delayed emergency services since the Fire Service have worked out there is a barrier on Staveley Road and Turnham Green Terrace reopened (if you are counting that as an LTN).The cycleway, which is obviously not an LTN, clearly did cause delays to emergency service vehicles in the early stages of its implementation but, since they have avoided logjam in the centre of Chiswick by kettling vehicles at each end, I don't think there have been significant extra delays to emergency vehicles on the High Road.

Francis Rowe ● 1993d

Ed, I do feel safer with cars around and thought it was only me, obviously not, that's why I travel so far to shop.

Janice Evans ● 1993d

You’ve  ignored the previous work that the authors reference. There is a spike in reports of LTN infrastructure delaying emergency services when it is first introduced.  However, over time although the infrastructure doesn’t change, the reports of it causing delays drops. That can only be explained by over-reporting of the cause of delays when the LTN is first introduced.And you are also completely overlooking the critical point, that no matter what the reported cause of the delay, the actual delay in terms of time is no higher in an LTN than the area around an LTN or elsewhere: they are already neutral!

Tom Pike ● 1993d

Suppose like everything we won't know until we need an Ambulane or Fire Service.

Janice Evans ● 1992d

Ed -As you are someone who continues to work from home, perhaps you'd appreciate a perspective from someone who has had little choice but to remain out and about.Now with markets back up and running in central London, I am repeatedly on my old route which I've travelled numerous times over the past 8 years - mainly along Euston Road and further into Islington.The traffic is now worse. It is not an aberration, it is a constant. It is not down to volume.The lights as you come down from the Paddington overpass are poorly sequenced so traffic backs up often to the previous exit for the west end. As you idle, you get to look at the offices to the right. They are largely still vacant. Easy to spot because the lights aren't on. Hotels are not yet operational which is also evidenced by the vacant Hilton. Again, no lights on and all curtains opened identically. Admittedly this isn't on TfL or UCL letterhead, but it's also not rocket science so even a cancelled guy like me can sort this one out. Once you get past the Magistrates Court, things ease up until you get to Great Portland Street. Madame Tussaud's remains closed so there are no tour buses or hordes of pedestrians pushing buttons so they can cross. But at GP Street, that's where the unused temporary cycle lane starts. To be fair, it gets used by motorcycles. As you idle there you will also see vacant offices and things like Pret still closed. Again, no hotels are up and running bar the one by the library which has lights on but no guests - not a one - in the ground floor cafe. As you near Kings Cross, the traffic completely disappears. Pre-pandemic this would never be the case. There would be two lanes of cars/lorries and the bus lane all slowly crawling east. Now, with just one lane, it sails because there are fewer vehicles. And then you get to the left turn in front of KX Station. That used to be a nightmare. Now it is empty. Pin-drop empty. No tour buses, no Ubers just the idling black cabs in front of the Great Northern Hotel. Once Eurostar is delivering passengers and tourists return not to mention offices filling back up (if they ever do) and delivery vehicles increasing to satisfy increased demand for services across all sectors, then we might come close to pre-pandemic traffic levels. But the big "if" of offices being used as they were is pretty much not going to happen. Follow any commercial real estate research and it all says the same thing: the work week as we knew it is gone and so, too, is the way office space gets used. How that space will evolve will take years to emerge before we can begin to understand how it will impact any sort of travel infrastructure.

David Lesniak ● 1992d

Or to Francis' point, the LTNs may have delayed emergency vehicles when they were first introduced, but once they knew they were there, they updated their routes to bypass them, which might explain why reports of LTN delays declined.That might not have impacted times too much because traffic volumes may have been down during the introduction phase, hence why reports of delays due to traffic also declined.Of course not all LTNs are physical barriers - some are just cameras - which emergency vehicles can ignore and there should be less traffic congestion inside the LTN which might mitigate any delay outside.The concern would be more traffic congestion on main roads outside the LTNs as traffic is directed away from LTNs.  If that happens you might see a spike in "traffic" delay coding which in a non-Jesuitical way, some may want reclassified as indirect LTN delays.

Ed Saper ● 1992d

Obviously a different outlook, but I see your point :-)

Ed Saper ● 1992d

Francis appears to have misread the paper when he wrote "The first question that should be asked is not 'do LTNs delay emergency vehicles?', they clearly do."The actual delays are in Table 1 and do not show the delays in the LTNs he claims, but rather the opposite. Here are the reductions in response times for the first fire engine before and after the introduction of the LTNs :Inside LTNs: 14±2sLTN boundary roads: 13±2 sSurrounding areas: 10±2 sOutside LTNs: 11±1 sThe reduction in the time for the second fire engine to arrive at an incident was more pronounced:Inside LTNs: 21±3sLTN boundary roads: 14±3 sSurrounding areas: 10±2 sOutside LTNs: 13±1 sIn both cases there was a larger improvement in response time inside the LTNs compared to outside them.The first question that should be asked is 'do LTNs delay emergency vehicles?', and the answer is they clearly don't.

Tom Pike ● 1992d

Tom, as another poster put it recently, you are just creating obfuscation about what is a very simple issue to make a case for the defence.

Francis Rowe ● 1992d

No, you simply misread what the paper said, which is "these findings add to evidence that LTNs do not adversely affect emergency response times."That evidence directly contradicts the claim you made after you had read the paper: "'do LTNs delay emergency vehicles?', they clearly do."

Tom Pike ● 1992d

Thanks for the perspective.  I had a look last night and traffic levels are roughly 90% of pre-pandemic levels - on certain days 100% or above.  That's UK wide and I don't dispute that central London roads are emptier - which must be nice!In London, the interesting thing is that Tubes are still 60% down (at one point they were operating at 5% capacity) and buses 40% down.A lot of the people who would be in central London would be coming in by public transport - and they're not.I agree - the old ways are gone.  The bottom has fallen out of the office market - but industrial / warehouse is booming. Although I'm technically working from home still, I had a meeting in Chiswick and then catchups at the office and now back at home.A lot of people are going to be spending a lot more time locally - and the money they used to spend on commuting and city-centre sandwiches, coffees, drinks etc. may be spent locally instead.  "Working from home" doesn't always mean staying all day at home.  We do lots of meetings around Chiswick - in coffee shops, the business park and more formal ones at The Clayton.  Ozzie - your competition down the Gunnersbury end - is turning his car park into a garden (only used for deliveries - not open to customers) and renovating the basement - great places for meetings, phone calls and team chats.Granted there will be losers in central London, but opportunities for local retailers.

Ed Saper ● 1992d

How is a delay defined?  Do they still put in a delay code even if they hit the target response time?Is it possible that some were delayed by LTNs (or at the very least, the delay was attributed to an LTN), while overall average response times still fell?

Ed Saper ● 1992d

A delay due to LTNs would lead to a comparative increase in the emergency response time for LTNs compared to other areas. In fact as the paper showed there was a comparative decrease in the response time for LTNs.The delay code is an assessment of what might be the cause of a delayed response time, and is used in a minority of all incidents. There was an increase in the traffic-calming code in LTNs, but that was more than offset by a decrease in traffic/roadworks code:https://findingspress.org/article/23568-the-impact-of-2020-low-traffic-neighbourhoods-on-fire-service-emergency-response-times-in-london-uk?attachment_id=59395It is quite possible that LTN's, as is their intent, led to a drop in traffic inside them that more than offset any delay from the fire engines having to negotiate any physical modal filter. However, the biggest improvement in response times was found for LTN's using physical barriers rather than cameras, which might reflect that physical barriers produce a higher level of compliance leading to lower traffic levels and hence faster overall response times in LTNs. Overall the evidence from this paper suggests that measures that reduce traffic lead to an improvement in emergency response times, even taking into account any delay in negotiating the barriers used to reduce the traffic.

Tom Pike ● 1992d

As I said, the article is completely useless.  By comparing a pre-pandemic period with a pandemic period and not making any assessment of traffic volumes or statistical significance the entire analysis is fundamentally flawed - all they have shown is a correlation not a causation.The data in the study showed that first engine response time post introduction of LTNs was 4.8% less in LTNs, 4.8% less in the boundary roads of the LTN but only 3.4% less within 500m of the LTNs and 4.2% more than 500m from an LTN.  Is the difference between 4.8% and 4.2% reduction in emergency response time statistically significant?  Within 500m of an LTN the reduction in emergency response time was 3.4% again is that statistically significant and if it is what is the cause of that lower reduction - a perfectly logical conclusion from the data is that LTNs actually increase emergency response times in the areas closest to an LTN.

Justin Stephenson ● 1992d

And if the study reported a 10% increase in delays due to LTNs you’d also reject that one out of hand as well?Or would you shout it from the rooftops because it confirmed what you already knew?

Ed Saper ● 1992d

Justin, the article deals comprehensively with the statistical significance of the analysis. The tables give the standard errors, which give estimates of the uncertainty in the delays, and the p-value which gives the probability that the results could occur by chance for all the comparisons. You are asking questions that are answered in full in the paper, so I hope you are unfamiliar with basic statistics rather than being disingenuous in asking these questions. You've also miscalculated the delay outside of the LTN's but that's less of a concern than your misunderstanding of statistical significance.The answer, to repeat what is clearly stated in the paper, is that there is no statistically significant comparative delay in and around LTNs or in the surrounding area compared to everywhere else. In statistical terms, with the p-values that are given, there is a no reason to reject the null hypothesis that LTN's cause no delays to the emergency services.

Tom Pike ● 1992d

Let us just say that we can agree to disagree because I used the actual data from the paper.  There is no analysis of traffic volumes and how they effected what I fully accept is a reduction in emergency response times - both in LTNs and outside of LTNs.The problem with quoting p-values (which I agree provide no reason to reject the null hypothesis) is that those are only a fair reflection if you have correctly taken into account all factors including traffic volumes and the reasons for changes in traffic volumes.  The web links you have provided show no evidence of those factors being considered properly - the authors appear, from your links, to have simply analysed the responses of emergency crews whereas what should have been done is to analyse actual traffic volume data around that time and any permanent or temporary reasons for changes in traffic volumes (for example road works or other events).

Justin Stephenson ● 1991d

Every reason to reject this in Ealing as NO repeat NO baseline traffic volume data was captured.  If there was no baseline then it follows that any delay (or not) is immeasurable.

N V Brooks ● 1991d

I have a feeling I have explained my criticisms of the study rather badly and with fat fingers on a calculator got one of the %age reductions wrong.Looking at the actual figures for reduction in response time pre & post LTN the study figures for first engine areInside LTN : pre LTN 294 sec, post 280 (a 4.8% reduction)LTN boundaries: pre LTN 273 sec post 260 (a 4.8% reduction)Surrounding roads (within 500m of LTN) pre LTN 296 sec, post 286 (a 3.4% reduction)Outside 500m of LTN pre LTN 313 sec post LTN 303 (a 3.2% reduction).A plausible explanation is that the difference in rates of reduction in response time between the LTN and its boundaries compared to further afield is that the introduction of LTNs reduced traffic volumes within the LTNs by more than further afield areas such that even if some aspects of LTNs increase emergency vehicle response times the positive effects of reduced traffic volumes outweighed this so that the net effect was a clear benefit to LTNs.That is certainly the explanation that the authors of the study maintain is correct and it is most definitely a plausible explanation.But here is the problem.  We know that the introduction of LTNs change traffic patterns - that is part of the purpose of LTNs.  What we do not know for the recent LTNs is how those changes will impact on the wider road network.  We know from studies in the past that the likelihood is that there will be short term effects and probably rather different long term effects.  So if the introduction of LTNs has displaced a lot of traffic onto other roads what we should expect to see is traffic volumes rising on those other roads (there may be/will be some traffic evaporation) and we know that traffic volumes have an effect on emergency response times.  We also know that traffic displacement is not merely onto roads close to the original routes but can be onto routes that are several miles away from an LTN.Of course the problem is compounded by the fact the authors were studying the introduction of LTNs during a pandemic which has had a severe effect on traffic volumes but also a very complicated effect.In very simplistic terms what we do not know is to what effect the reduction in response times outside of the LTNs has been adversely effected by traffic displacement from LTNs.The reason this is important is that it means we cannot establish a base line for the amount of reduction in response times caused solely by the pandemic reducing traffic volumes - and we need that in order to isolate the additional effects the introduction of LTNs have caused.The difference between a 4.8% reduction in response time within an LTN or on its boundary compared to a 3.2% further afield ought to be significant, there is a 50% difference between the 2 figures but that judgment might be very different if after taking into account traffic displacement the differences dramatically reduced.The point being that if there was a dramatic reduction in the differences between response times following taking into account traffic displacement that might result in the conclusion that LTNs benefit emergency response time within the LTNs and its immediate boundary roads but at the expense of worsening response time elsewhere.I should emphasise that these are issues which do need to be researched and it may well be the case that further research ultimately confirms that overall LTNs are a benefit to response times both within LTN areas and further afield. And just to annoy many of the anti LTNs/cycleway posters here, I believe that the LTNs will need to stay in place for at least a couple of years for there to be sufficient data to enable the research to be robust

Justin Stephenson ● 1991d

Thanks Justin for your take on this.We should take this study at face value - during a pandemic when traffic volumes were substantially lower at times, LTN's didn't delay emergency vehicles.It is plausible that if traffic is redirected onto other roads and overall traffic increases, there may be increases in times, even if they can't be directly attributed via coding.It is also plausible that there may be mitigating effects - traffic inside LTNs might be lower, barriers may be replaced or upgraded to allow emergency vehicle access, routes are updated and traffic may evaporate.This study doesn't predict the future and, as Justin says, further data will need to be collected to discover what the impact actually is.

Ed Saper ● 1991d

Dubious claims galore are being made from these number crunching studies. They really don’t tell us much.Btw.. Is Rachel Aldred the same person who has written about Walter Benjamin? If so, she should know better.

Russell Pearson ● 1991d

The statistical analysis in these two papers follows the same "number crunching" approach employed to test the efficacy of a vaccine: compare the outcomes from a group with the treatment and a control group and see if there is a statistically significant difference.Thankfully, not many people would claim that vaccine trials "really don't tell us much."

Tom Pike ● 1991d

So are you seriously suggesting that these sociology studies follow the same process and rigour of medical science? Even the authors state they should be treated with ‘caution’.

Russell Pearson ● 1991d

The statistical approaches are the same, and medical studies often use exactly the same wording.These are both far more rigorous studies than the flawed reports in newspapers that you have been previously willing to promote.

Tom Pike ● 1991d

They do scrape ahead of journalism. As for promoting them as a gold standard of scientific method just because they use the same words and statistical approach as other more trusted fields, well that would be very naive indeed. Makes me wonder how aware you are, of just how problematic this whole area is, and not just for ideologically driven social science reports.

Russell Pearson ● 1991d

The very same who wrote about him in 'Socialist Review'. You do realise however that any criticism of the High Priestess of "the hypothetical facts that suit our narrative" is not tolerated.  Cue another stream of dubious 'facts' in support of the nonsensical 'you don't like LTNs therefore you must hate cyclists' position.

N V Brooks ● 1991d

If the flaws are so obvious, maybe you could point them out? If you can't then it looks like empty disparagement of a peer-reviewed article on the basis of a dislike of its conclusions.

Tom Pike ● 1991d

I wonder if Labour Party entryism as almost perfected by the SWP is at play. Cycling Trots?This we can agree on:“these analyses of mid-pandemic crime data should be treated with some caution”

Russell Pearson ● 1991d