History, Risk & Infrastructure

This article provides historical context on the contrasting histories of cycling in the Netherlands and Britain.

It is a slightly edited version of a paper of mine published in the Journal of Transport and Health in 2014. It has therefore been through a formal peer-review process. As I own the copyright, the original paper may be downloaded as a PDF from here. The theme of the paper is to compare trends in bicycle use and risk in the United Kingdom and the Netherlands during the period since 1950. The two countries have pursued more or less diametrically opposite bicycling policies, until recently.

Please note that the papers referenced from the 2013 International Cycling Safety Conference are no longer available at the ICSC web site for some reason.


Executive Summary

Cycling has consistently been safer in the Netherlands than the UK. Nevertheless, safety has improved in both countries over time. Between 1980 and 2011, the cyclists’ fatality rate declined by 67% in the Netherlands and 57% in the UK. Per capita bicycle use was sustained in the Netherlands throughout the post-World War Two era, peaking in the early 1960’s and only declining for a decade before recovering. UK bicycle use peaked in 1952 and declined permanently. The survival of popular bicycling in the Netherlands through the 1950’s and 1960’s was fundamental to the development of effective bicycling policies after the 1970’s. The Dutch network of cycle tracks and routes increased from 9,000 km in the mid 1970’s to approximately 29,000 km currently. The annual distance cycled per capita increased by 30% in the ten years to 1988, but has not materially increased since then. In the UK, cycling has a long heritage as a marginalised form of travel. This continues to hinder efforts to achieve a national cycling revival. Cycling on quiet urban and rural roads in the UK incurs much lower risks than the national average fatality rate would suggest. Networks that enabled cyclists to avoid main roads, especially rural A-roads, could provide safety levels comparable to the Netherlands and Denmark. There are towns in the UK with segregated cycling networks, but few cyclists. This is because a range of measures must be invoked to achieve large modal shifts to cycling. Local authority support is a critical factor.

Introduction

In November of 2013, I travelled to the town of Helmond in the Netherlands to attend the International Cycling Safety Conference (see http://icsc2013.blogspot.nl/), and present a paper on modal risk (Wardlaw, 2013) [1].

The trip was an unexpected immersion into the debate about infrastructure for cyclists. First to speak was the famous advocate of Vehicular Bicycling, the American John Forester, who has long maintained that bicyclists fare best when they ride competently with motor traffic (Forester, 2013)[2]. The following speaker was Mr Tom Godefrooij of the Dutch Bicycling Embassy (Godefrooij, 2013)[3], who emphasised the much broader appeal of segregated cycling, especially for children, women and elderly people. Preference for segregation has been further supported by results from China, reported in this journal (Lusk et al, 2014)[4].

It would be hard to dispute the segregation case from the visible evidence around Helmond, or Amsterdam, which I visited before the flight home. Cyclists pour down the lanes and cycle tracks, queues back up into the distance at traffic lights: young women, old men, children, businessmen, girls and boys going out on the town. Barriers to cycling are apparently low in the Netherlands.

For any advocate of mass bicycle use, the Netherlands stimulate envy and curiosity. Amsterdam is not that different from many British towns and cities; it is flat, its central area was built before the motoring era and it rains a lot. Yet in the UK, only Cambridge and Oxford (respectively 33% and 19% share cycle commuting) (Goodman, 2013)[5] have cycling levels comparable to what is typical in the Netherlands. My curiosity settled on these five questions:

  1. What are the histories of cycling in NL and UK?
  2. Are NL cyclists safer than UK cyclists?
  3. Does infrastructure affect safety? (Mr Godefrooij was neutral on this point).
  4. How does the UK lack of infrastructure affect safety?
  5. Is lack of UK cycling just due to lack of infrastructure?

There was no existing single paper that addressed all these points. This Viewpoint seeks to fill this gap, based on readily available evidence.

Materials and Methods

Evidence was gathered primarily from reviews, reports or books, mostly published in the last five years, notably:

  • Bicycling, Health and Safety (OECD/International Transport Forum, 2013); The International Transport Forum at the OECD describes itself as “an intergovernmental organisation with 54 member countries. It acts as a strategic think tank for transport”.
  • A Handbook of Road Safety Measures (Elvik et al., 2009); 1000-page compendium of global road safety research.
  • City Cycling, (Pucher and Buehler2012). (relevant chapters are cited individually); This book presents a wealth of recent international research into the factors leading to successful cycling programmes.
  • Infrastructure, Programs, and Policies to Increase Bicycling: An International Review (Pucher et al. 2012)[6]. This review describes fourteen successful city cycling programmes.
  • The Dutch Bicycle Master Plan (Welleman, 1999)[7]. A comprehensive history of cycle use and policies to support cycling in the Netherlands.
  • Policies for Promoting Walking and Bicycling in England: a View from the Street (Pooley et al., 2013)[8]. A qualitative investigation in four English towns of current transport choices and how to reduce barriers to cycling.
  • Interventions to Promote Cycling: Systematic Review (Yang et al, 2010)[9]. A systematic literature review identifying the best quality studies out of nearly 28,000 documents screened.

In addition, data on personal transport and road casualties were gathered from national records of the UK and the Netherlands (hereafter “NL”), taking 1950 as a starting point. Earlier data are not readily available. Confidence intervals were not consistently available, and they have not been included. Comment has been made where uncertainty is large. The nomenclature used for cycling infrastructure is as follows:

  • Cycle route: off-highway right of way
  • Cycle track: segregated way parallel to a road
  • Cycle lane: unsegregated lane painted on the road

Speed reduction is included within the scope of “infrastructure”.

Review of Evidence

Historical Review since 1950, NL and UK

NL and UK Cycling (Early Post-World War Two Era)

Between the end of the Second World War and the first oil crisis in 1973, both NL and UK experienced increases in car ownership, increases in road deaths, and declines in cycle use.

Car ownership in NL was only 14 per 1,000 capita in 1950 (Wellman, 1999, p27), as against 40 per 1,000 capita in the UK (Gov.uk, 2014)[10]. However, by 1970, NL car ownership had achieved parity with UK, at just over 200 cars and light trucks per 1,000 population (Pucher and Buehler, 2012a Fig. 2.5)[11]. Perhaps linked to this rapid motorisation, road deaths in NL increased more rapidly than in the UK during these two decades, as Chart 1 shows:

Chart 1

Chart 1: Per Capita Fatality Rates NL and UK 1950-70 (All types of road users)

Data sources: NL, Centraal Bureau voor de Statistiek; UK, Department for Transport;

The patterns of bicycle use in NL and UK country were different in this period:

Chart 2

Chart 2: Distance Cycled per Capita in NL and UK, 1950-1980

Note: NL data from Wellman, 1999 p28 and Centraal Bureau voor de Statistiek, UK data from Department for Transport and Office of National Statistics.

Despite the rapid growth in NL car ownership from 1950, cycle use also continued to grow. Per capita annual distance cycled peaked in the early 1960’s at almost 1,500 km and then declined, recovering in the late 1970’s. At this nadir, cycling still made up 30% of non-pedestrian trips in Amsterdam, 35% in Eindhoven and almost 50% in  Enschede. In contrast, the modal share was 4% in Manchester (Welleman, 1999, Fig. 10).UK cycle use peaked in 1952 at 465 km annually per capita, declining to 70 km annually per capita by 1970. There followed a modest recovery after 1973.

NL and UK Cycling (Later Post-World War Two Era)

In the mid 1970’s, the long decline of cycling in Amsterdam, and a few other Dutch cities, reversed; cycle use started to grow again. Outrage at the scale of road deaths, especially of children, the oil crisis of 1973, and a general sense that enough had been conceded to the motor car, were factors bringing about a change in public attitudes. This in turn led to government action. Bicycling infrastructure programmes began to attract significant official support. The Netherlands already had a legacy of about 9,000 km of cycle tracks and paths. This increased to 16,000 km by 1988, following central government funding (Welleman, 1999). By 2009, this had increased to 29,000 km. For nearly forty years, the Netherlands has coordinated land development, public transport, and restrictions on car use, to support its cycle infrastructure programmes. Per capita cycle use grew by 30% after the late 1970’s to approximately 880km per capita annually (NL Centraal Bureau voor de Statistiek). Chart 3 shows that since the late 1980’s, NL per capita bicycle use has plateaued.

In the UK, there was also a cycling revival after 1970, but it was from an extremely low level of use; from 70 km/annum to a peak of 117 km/annum in 1984. This revival was ignored by successive governments and reversed after the mid-1980’s, during a period when government policy was focused on road building and increasing car ownership. By the mid-1990s, cycle use had been in decline again for a decade. The UK government finally launched the National Cycling Strategy in 1996. It was not supported by funding or local authority interest and was in effect abandoned by the subsequent government. Despite this lack of central support, a few UK cities have achieved growth in cycling, usually due to alliance of campaigners with the local authority. Edinburgh is a good example (Spokes, 2014)[12].

Chart 3

Chart 3: Distance Cycled per Capita in NL and UK, 1980-2010

Note: NL data from Centraal Bureau voor de Statistiek, UK data from Department for Transport and Office of National Statistics

In 2005, UK government funding supported the establishment of six “Cycling Demonstration Towns” by Cycling England. This programme of promotion, combined with better infrastructure, ran from 2005 and was expanded to include a further twelve towns or cities in 2008. The programme ended in 2011 with the abolition of Cycling England. As of 2013, new funding has been awarded to eight “Cycle City Ambition” winning cities, few of which were previously funded through Cycling England. This intermittent funding has had little effect on UK cycling levels. Analysis of national Census data for England and Wales (Goodman, 2013) shows that cycling for commuting is still in slow decline, except in a few towns and cities where local authorities are pursuing pro-cycle programmes: notably London, Bristol, Oxford, Cambridge and Brighton.

Risk in Cycling

Fatality Rates in NL and UK 1980 to 2011

It is possible to compare cyclist fatality rates in NL and UK. Both countries have databases of national distance cycled. Similarly, both countries hold long-term records of cyclist fatalities.

Chart 4

Chart 4: Bicyclist Fatality Rates per Billion KM 1980-2011, NL and UK

Note: Year 0 = 1980; Excel’s Exponential Trend facility does not work with dated years,

The average risk in the UK has remained consistently about twice that for NL. Nevertheless, the improvement across time has been substantial in both countries. The Exponential Trend facility of Excel has been used to provide best-fit equations, and these equations were used to calculate the percentage improvement in each country. The results are:

NL      67% decline in fatality rate 1980-2011

UK      57% decline in fatality rate 1980-2011

If all road deaths fell by more in one country than the other, then this would probably affect the outcome for cyclists. It turns out that road deaths fell by almost the same percentage in both countries (taking three year averages 1980/82 and 2009/2011): by 65% in NL and by 66% in UK.

Risk by Age Group in NL and UK

Bicyclist fatality rates by age group are available for 2008 for NL (de Hartog et al, 2010)[13] and England (Mindell et al, 2012)[14]. The age groups are not identical in both sources (i.e. NL child is <15 y.o., England child is <17y.o.; “>70 y.o.” is one age group in English data). English data are averaged from 2007-2009. The results are shown in Chart 5.

Chart 5

Chart 5: Bicyclist Fatality Rates by Age Group in NL and England, 2008

The results show that the safety advantage for NL cyclists is not spread evenly across all age groups, but is greatest for young riders and declines with age, until there is little material difference for the 70’s and older age groups. For riders younger than 50 y.o., the advantage is greater than the all-ages averages would suggest, the risk being typically five times less than for the equivalent age groups in England. Males account for 80% of UK distance cycled, and 85% of fatalities (and 90% of child fatalities). Males also tend to have higher fatality rates than females (Mindell et al, 2012), by a factor of 1.5-2. In NL, females make up just over half of the cycling population. This preponderance of males causes UK cycling risk to appear worse than it actually is.

Infrastructure Effect on Cycling

There is an abundance of research into the effects of infrastructure on safety and participation. Unfortunately, only a small proportion of it is based on observations in NL. Therefore any comment on the effect of Dutch infrastructure programmes will have to be inferred from the broader range of experience. The evidence has been subject to reviews, and in some cases meta-analysis, in existing published documents. The various sources come to generally similar conclusions for both safety and participation.

Infrastructure and Safety

The conclusions of the International Transport Forum (OECD/International Transport Forum, 2013) are probably the most authoritative available:

  • Speed management is a critical and effective tool to reduce the severity of bicycle/motor vehicle crashes
  • Where speeds cannot be lowered, or where traffic densities are high, authorities should seek to separate bicycle and motor traffic.
  • Separated bicycle tracks are an attractive option as they generally produce fewer and less severe crashes in their linear sections – however, safety may be compromised at junctions, where crashes may increase unless specific counter-measures are undertaken
  • Crash risk at bicycle track/road interfaces is exacerbated by poor sight lines, and confusion regarding expectations of cyclists vis-a-vis motorists and vice-versa. Proper design of junctions contributes to lower crash risk.

A number of other reviews reached similar conclusions. The Danish Road Administration (1994[15], 1996[16]), the Dutch Institute for Road Safety Research (SWOV, 1994)[17] and the UK Transport Research Laboratory (2011)[18] all concluded that cycle tracks improved safety on links but increased crash rates at junctions. The overall effect on safety was concluded to be neutral. The increased risk at junctions is most likely due to the separation, which leads drivers to be less aware of cyclists. Also, segregated cyclists must scan a greater range of view (c. 240 degrees versus 45 degrees) when approaching a junction (Franklin, 2002)[19].

The Danish Safety Administration advised that cycle tracks were most appropriate when motor traffic speeds and volumes were high. This is supported by recent Danish research for rural cycle tracks, showing up to 80% decrease in fatalities (Jensen et al., 2010)[20].

Lack of Infrastructure and Safety: UK versus NL and Danish Cyclists

UK data (Department for Transport, 2013)[21] are available for cyclist fatality rates by road class (Table 1). These show risks of cycling with motor traffic in different speed environments. The Department declined to provide confidence intervals.

Table 1: UK speed limits by road class
Rural A-road (Dual carriageway) 70 mph (110 km/h)
(Single carriageway) 60 mph (100 km/h)
Urban A-road 30 or 40 mph (50-65 km/h). A few arterials: 50 mph (80 km/h)
Rural B & C roads Formally 60 mph (100 km/h) but as such roads tend to be narrow and twisting, traffic speeds are usually 20-40 mph (30-70km/h)
Urban B & C roads 20 or 30 mph (30-50 km/h)

In the UK, dual carriageways are similar to motorways, although cyclists are often allowed to use them. Many other A-roads are traditional, meandering routes not much different from B-roads, where traffic speeds are lower.

The following chart shows the results for cyclists, averaged for the two years, with UK pedestrian, NL and DK national cyclist fatality rates for comparison.

DfT Risk by Road Class 2011 2012

Chart 6: Risk in UK Bicycling by Road Class, 2011-2012, with Comparisons

Data Sources: All UK data from Department for Transport, NL Centraal Bureau voor de Statistiek, DK Danish Ministry of Transport. Confidence intervals not available. UK cyclist and pedestrian data include fatalities in falls.

Cycling on UK rural A-roads carried more than eight times the national average risk of fatality per KM travelled (Note: more recent data show lower risks). In contrast, UK cycling on urban roads bore risks similar to cycling in NL or DK. This is a low level of risk, being about half the risk per KM travelled of UK walking. Cycling on rural B and C class roads also carried low risks, although higher than the national averages for NL and DK. The risk in UK cycling on urban A-roads was about double the national UK average, although only a little higher than the national average for UK pedestrians.

Infrastructure Effects on Bicycle Use

Social research finds that the majority of the population is “traffic sensitive”, that is, people fear riding a bicycle in traffic. Perceived traffic danger is a significant deterrent to mass cycle use (Pooley et al., 2013; Jacobsen et al., 2009[22]). High income countries without extensive cycling infrastructure have low levels of cycling and in particular, low involvements of children, women and elderly people (Pucher and Buehler, 2012a). Analysis of modal share in England and Wales reveals cycle commuting is growing only in a few towns and cities (Goodman, 2013), mostly ones with strong campaigning groups and local authority commitment to cycling. Infrastructure investment is part of a range of measures being applied in these places.

Surprisingly, building cycle infrastructure in isolation does not necessarily have much effect. In the UK, there are a number of “new towns” built from scratch with good quality cycling networks separate from the road system. These include Milton Keynes, Stevenage, and Livingston. The network of Stevenage is of particularly high standard. Yet today it is neglected, and cycle use is low at 3% share of commuter trips (ECF, 2013)[23]. Early Dutch programmes mirrored this experience. Construction of networks in The Hague, Tilburg and Delft between 1975 and 1987 attracted existing cyclists but did not increase cycle use much overall (Wellman, 1999, p43-44).

A literature review of interventions to promote cycling (Yang et al, 2010) screened almost 28,000 references, but found only a single good quality study (i.e. with a controlled comparison) reporting on infrastructure alone as an intervention to promote cycling. This concerned a programme in Delft, NL, to improve the connectivity of cycle routes on one area of the city (Wilmink & Hartman, 1987)[24]. The results showed a slight increase (2% absolute and 5% relative) in household trips by cycle over a three year period. This increase was from an already high level of 40% of trips.

Can cycling recover in the absence of infrastructure investments? Charts 3 and 4 reveal a UK cycling recovery of almost +70% in the ten years to 1984. This revival has never been studied, therefore any identifiable causal factors are not known. Such factors might be relevant today.

Discussion and Conclusions

A previous Chief Medical Officer for England called for an eight-fold increase in cycling to address the public health consequences of low physical activity levels (Donaldson, 2010)[25]. This reinforced earlier recommendations for a major cycling revival, notably the House of Commons report on Obesity (Select Committee on Health, 2004)[26]:

“If the Government were to achieve its target of trebling cycling in the period 2000-2010 (and there are very few signs that it will) that might achieve more in the fight against obesity than any individual measure we recommend within this report.”

Has this historical review of cycling in NL and UK since 1950 provided insight into achieving the eight-fold increase?

Some lessons from history have to be acknowledged. The Dutch did not abandon the bicycle and then revert to being a cycling nation. Cycling in NL increased for many years after the end of the Second World War, and only suffered decline for about a decade between the mid 1960’s and mid 1970’s before recovering. The recovery had support across Dutch society and enjoyed government investment enduring decades, integrated with land planning, restrictions on car use and legal protection for cyclists, notably Strict Liability, which applies in NL but not in UK (Cycle Law Scotland, 2012)[27]. The relevant Dutch legislation is Article 185 of the Wegenverkeerswet (Road Traffic Act) 1994. It is likely that the steep increase in NL road deaths during the 1950’s and 1960’s created a revulsion against unfettered motor car use. This, combined with the surviving popularity of cycling, enabled broad political support for a range of measures to increase cycling, infrastructure included.

The response to traffic danger in UK was to restrict – indeed, virtually end – the independence of children (Hillman et al., 1990)[28].

From the history, the high level of NL cycling still extant in the early 1970’s was a key element in the creation of the bicycling networks. The average Dutch citizen cycles 30% more km per year now than in the early 1970’s. It is exceptionally difficult to dispute that infrastructure has enabled participation to grow from already high levels amid rising traffic levels. However, Chart 4 shows that per NL capita cycle use has not grown further since the late 1980’s.

It is also hasty to assume that building infrastructure alone will bring about a cycling renaissance from the low levels typical of the UK. Successful cycling programmes have been multifaceted. Essential elements have included (Pucher et al, 2012, Pucher and Buhler, 2012b):

  • Safe cycle parking
  • Integration with public transport
  • Education of cyclists and motorists
  • Legal protection for safe road users, such as Strict Liability laws
  • Bike sharing programmes
  • Restrictions on car ownership and use
  • Compact land development to shorted trip lengths
  • Time – to build networks and change public behaviour takes decades.

The example of Stevenage, UK is a good illustration of why the multifaceted approach is required. With its excellent infrastructure, Stevenage could join Oxford and Cambridge in being a cycling community. It does not, most probably for two reasons: the local council has no interest in promoting cycling; and, being a town built for driving, the grid of boulevards and ample parking makes the car all too convenient (Reid, 2013)[29].

The evidence presented in this viewpoint does contain findings that could be applied in promoting cycling in the UK, and perhaps other countries with low use:

1) The risks of cycling in UK are low if one avoids rural A-roads and minimises use of urban A-roads (Chart 6). It should be feasible for most towns and cities to build up a network of lesser streets, and off-highway routes through parks, that allow cyclists to avoid main roads. In addition, suburban areas could be subject to a 20 mph limit by default. Authorities should not be deterred by the flawed perception that urban cycling is dangerous. This is a key message for all those seeking an increase in active travel (with a caveat about the high risks of active travel for the elderly).

2) The evidence shows cycle tracks improve safety where traffic speeds are high and junctions are far apart. Cycle tracks would make a significant contribution installed as standard along all new by-passes, urban arterial routes, and rural A-roads where this would provide important links.

3) Cycle tracks are unlikely to be a success if they are installed along roads with many driveways and side roads. Dutch urban areas have been developed so that main roads generally do not have driveways or lanes emerging onto them (International Transport Forum, 2013). Where this would be a problem, as in suburban streets or old town centres, the solution is to allow sharing of road space with calmed traffic. This should be done in UK too.

4) General road safety improvements make cycling safer. The safety record for UK cyclists is poorer than for NL cyclists, and for children it is scandalously worse (Chart 5), but there has been considerable improvement over time. The decline in fatality rates 1980-2011 was 67% for NL cyclists and 57% for UK cyclists (Chart 4). Proponents of active travel should be confident to expect even larger reductions in risk, if a competent infrastructure programme is progressed, complemented by the other essential elements already listed.

5) Cycling must enjoy institutional respect. Many UK cycle tracks are narrow and badly paved, and force users to give way at driveways and side roads (Franklin, 2002). In many cases, pavements (sidewalks in North America) have simply been re-badged for cyclists to use, without modification. This shoddiness of execution is probably the biggest risk to infrastructure being accepted by existing cyclists, let alone by the eight-fold wave of newcomers that has been called for.


References

[1] Wardlaw, M. 2013. Like for Like Risk Assessement  of Road safety by Travel Mode. Paper presented at the International Bicycling Safety Conference, Helmond, Netherlands.

http://www.icsc2013.com/papers/mindell2013_Risk%20assessement%20by%20travel%20mode.pdf (accessed 18/5/14)

[2] Forester, John. 2013. Two Systems for Bicycle Operation: Obeying the rules of the road, or Cyclist Inferiority; with some discussion of the Dutch system. Paper presented at the International Bicycling Safety Conference, Helmond, Netherlands.

http://www.icsc2013.com/papers/Forester2013_Two%20Operating%20Systems.pdf (accessed 18/5/14)

[3] Godefrooij, Tom. 2013. Why the so called ‘vehicular bicycling’ concept is creating a false dichotomy. Paper presented at the International Bicycling Safety Conference, Helmond, Netherlands.

http://www.icsc2013.com/papers/Godefrooij2013%20vehicular%20bicycling%20false%20dichotomy.pdf (accessed 18/5/14)

[4] Lusk, Anne. 2014. Gender and used/preferred differences of bicycle routes, parking, intersection signals, and bicycle type: Professional middle class preferences in Hangzhou, China. Journal of Transport & Health 1(2):124-133

Doi: 10.1016/j.jth.2014.04.001

[5] Goodman, Anna. 2013. Walking, bicycling and driving to work in the English and Welsh 2011 Census: Trends, socio-economic patterning and relevance to travel behaviour in general. PLoS ONE 8(8):e71790.

Doi: 10.1371/journal.pone.0071790.

[6] Pucher, John, Jennifer Dill, and Ralph Buehler. 2012. Infrastructure, programs, and policies to increase bicycling: An international reviewPreventive Medicine 50:S106-S125.

Doi: 10.1016/j.ypmed.2009.07.028

[7].Welleman, Ton. 1999. Dutch Bicycle Master Plan. Directorate-General for Passenger Transport. The Hague.

[8] Pooley, Colin, Dave Horton, Griet Scheldeman, Caroline Mullen, Tim Jones, Miles Tight, Ann Jopson, Alison Chisholm. 2013. Policies for promoting walking and bicycling in England: a view from the streetTransport Policy, 27, 66-72.

Doi: 10.1016/j.tranpol.2013.01.003

[9] Yang, Lin, Shannon Sahlqvist, Alison McMinn, Simon Griffin, David Ogilvie. 2010. Interventions to promote cycling: systematic review. British Medical Journal 341:c5293.

[10] Gov.uk, 2014. “All Licensed Vehicles and New Registrations”. Web site of the UK Government

https://www.gov.uk/government/statistical-data-sets/veh01-vehicles-registered-for-the-first-time#table-veh0103 (accessed 22/07/14)

[11] Pucher, John, and Ralph Buehler. 2012a. International Overview: cycling trends in Western Europe, North America and Australia. Chapter 2 in: City Cycling, Pucher and Buehler. The MIT Press.

[12] Spokes (The West Lothian Cycling Campaign). 2014. http://www.spokes.org.uk/wordpress/about/

(accessed 3/8/14)

[13] De Hartog, Jeroen, Hanna Boogaard, Hans Nijland, and Gerard Hoek. 2010. Do the Health Benefits of Bicycling Outweigh the Risks? Environmental Health Perspectives 118(8):1109-1116.

[14] Mindell, Jenny, Debbie Leslie, Malcolm Wardlaw. 2012. Exposure-based, ‘Like for Like’ Assessment of Road Safety by Travel Mode Using Routine Health DataPLOS ONE 7(12):e50606

Doi: 10.1371/journal.pone.0050606.

[15] DANISH ROAD ADMINISTRATION (1994). Cyclists’ behaviour in urban roundabouts. Notat 13.Danish Road Administration. Denmark.

[16] DANISH ROAD ADMINISTRATION (1996a). Trafiksikkerheds-effekten af cykelbaner i byområder (The traffic safety effect of cycle lanes in urban areas). Rapport 50. Danish Road Administration. Denmark

[17] SWOV (1994). Towards a sustainable safe traffic system in the Netherlands. Research Activities No.1, March 1994, pp 8-9 & 16. SWOV. Leidschendam.

[18] Transport Research Laboratory. 2011. Infrastructure and Cyclists’ Safety. Report PPR580.

[19] Franklin, John. 2002. Achieving Cycle-Friendly Infrastructure. Presented to the Cycle-Friendly Infrastructure at the University of Nottingham.

http:/www.cyclecraft.co.uk/digest/cfi_jaf.pdf (accessed 18/5/14).

[20] Jensen, S.U., P Andersson, L Herrstedt. Handbook, Road Safety Effects, Road Engineering Measures. Danish Road Directorate, 2010.

[21] Department for Transport. 2013. Reported Road Casualties in Great Britain 2012. HMSO. London.

[22] Jacobsen, Peter, Francesca Racioppi, Harry Rutter. 2009. Who owns the roads? How motorised traffic discourages walking and bicycling. Injury Prevention 15:369-373.

[23] European Cyclist’s Federation. 2013. What Haopens When You Build it and They Don’t Come? http://www.ecf.com/news/what-happens-when-you-build-it-and-they-dont-come-the-cautionary-tale-of-stevenage/ (accessed 29/7/14)

[24] Winkman A, Hartman J, 1987. Evaluation of the Delft Bicycle Network Plan: final summary report. Ministry of Transport and Public Works, Netherlands, 1987.

[25] Donaldson, Sir Liam. 2010. On the State of Public Health: Annual Report of the Chief Medical Officer 2009. Department of Health, Crown Copyright, 2010.

http://webarchive.nationalarchives.gov.uk/20130107105354/http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/AnnualReports/DH_113912 (accessed 3/8/14)

[26] Select Committee on Health. 2004. Third Report (Obesity). The Stationery Office, 2004.

http://www.parliament.the-stationery-office.co.uk/pa/cm200304/cmselect/cmhealth/23/2302.htm (accessed 3/8/14)

[27] Cycle Law Scotland. 2012. Report on Strict Liability.

http://www.bicycling-accident compensation.co.uk/Research%20Report%20on%20Strict%20Liability%20for%20SG.pdf (accessed 18/5/14)

[28] Hillman, Mayer, John Adams, John Whitelegg. 1990. One False Move: a Study of Children’s Independent Mobility. Policy Studies Institute.

[29] Reid, C. The Sad Tale of a Cycle Network Innovatorhttp://www.roadswerenotbuiltforcars.com/stevenage/ (accessed 7/8/14)