Intended for healthcare professionals

CCBYNC Open access

Rapid response to:


The health risks and benefits of cycling in urban environments compared with car use: health impact assessment study

BMJ 2011; 343 doi: (Published 04 August 2011) Cite this as: BMJ 2011;343:d4521

Rapid Response:

Evaluating the benefits of public bicycle schemes needs to be undertaken carefully

This paper sets out to compare the health benefits of the Bicing
scheme (Barcelona's public bicycle share scheme) with possible risks
associated with increased bicycle riding. The key variables used by the
researchers include physical activity, exposure to air pollution and road
traffic injury. The authors rightly identify that although traffic
congestion is often a major motivator behind the establishment of public
bicycle share schemes (PBSS), the health benefits may well be the largest
single benefit of such schemes. Certainly PBSS appear to be one of the
most effective methods of increasing the number of bicycle trips across a
population, providing additional transport options and improving awareness
of the possibilities bicycles offer urban transport systems.

As noted by the authors Rojas-Rueda, de Nazelle, Tainio, &
Nieuwenhuijsen (2011), the number of PBSS have increased rapidly over
recent years, yet little work has been undertaken to evaluate their
impacts (Buttner et al., 2011; Shaheen, Zhang, Martin, & Guzman,
2011). Whilst Rojas-Rueda et al. (2011) should be commended for undertaking
what is understood to be amongst the first health impact assessment on a
PBSS, flawed assumptions regarding the proportion of Bicing users who have
substituted a motor vehicle journey invalidates their results. Put simply,
their assumption that between 90% and 100% of Bicing users would have made
the trip by car had it not been for the PBSS dramatically overstates the
actual rate of trip substitution. The available data demonstrates only
9.6% of Bicing trips are substituting for a car journey (City of
Barcelona, 2007). Moreover, 6.3% of trips taken on Bicing were previously
private bicycle journeys - neutralising any benefit to public health.

In addition to overstating the health benefits, this erroneous
assumption by Rojas-Rueda et al. (2011) has also led to overstated
environmental benefits of the Bicing scheme. For instance, the authors
state "As a result of journeys by Bicing, annual carbon dioxide emissions
were reduced by an estimated 9,062,344kg" (p. 1). Further: "Data on shifts
in mode of travel as a result of the Bicing initiative could not be found"
(p. 3). Data published by Anaya & Bea (2009), collected by the City of
Barcelona show users of the Bicing scheme to be substituting from other
modes of transport in the following proportions: Public transit 55.10%,
motor vehicle 9.60%, walking 26.10%, private bike 6.30% and
new trip 2.80%.

As the above figures illustrate, over one quarter of Bicing trips are
replacing pedestrian journeys, and given the evidence suggesting walking
trips have twice the health benefit of bicycle riding, on a per kilometre
basis (New Zealand Transport Agency, 2009), the results produced by Rojas-
Rueda, et al. (2011) likely overstate the health benefits of the scheme.

In addition to these aforementioned assumptions regarding the
proportion of Bicing trips replacing car journeys, the total number of
daily trips using Bicing also appears higher than what the available
evidence suggests. The authors calculate that there are 28,251 Bicing
members using the scheme daily and assume a 90% shifting from car
journeys. Assuming 28,251 members use the scheme daily and 90%
substituting for car use (despite evidence to the contrary), this equates
to 25,426 users substituting car for Bicing. Assuming these users take two
trips per day, 50,852 journeys would be made by these users. These
numbers, used by the authors to determine the health benefit of the Bicing
scheme conflict with the data made available by the operators (City of
Barcelona, 2011) suggesting something closer to 33,000 trips per day (6000
bikes used 5.5 times per day). Even this lower figure is quite possibly
higher than the actual figure, given that it is very rare for all 6000
bicycles to be in circulation at the one time.

Only those trips previously undertaken by car, as well as the 2.8% of
new trips generated by Bicing should be included in the physical activity
benefits of the Bicing scheme. One should also factor the health benefit
lost from the pedestrians opting for Bicing, given that the literature
widely regard walking to have twice the physical activity benefit of
cycling on a per kilometre basis (Fishman, Ker, Garrard, & Litman,
2011; New Zealand Transport Agency, 2009). Although these assumptions made
by the authors are the central flaw in the study approach, the sensitivity
analysis that assumed only 10% of Bicing trips were replacing a car trip
is much closer to the reality of the Bicing system, although there are no
reported calculations for this analysis.

The authors have assumed that 90% of Bicing users were new to cycling
when they signed up. If this was the case, it might be possible that these
new cyclists have a higher road traffic injury rate than regular cyclists,
yet this is omitted from the results and discussion, despite the authors
using road traffic injury as one of the key measures of health impact.

The carbon dioxide emission savings stated by the authors, previously
highlighted as artificially high, also fail to take into account the
redistribution and manufacture of the bicycles. Most PBSS use petrol or
diesel powered trucks to move their bicycles through the system and the
Bicing system is no different in this regard. The manufacture of 6000
bicycles, as well as their docking stations is not without significant
carbon dioxide emissions, and whilst it may well be less than the savings
as a consequence of reduced car use, it must still be factored into the

When calculating the health risks posed by road traffic injury, the
study only uses mortality and this fails to capture serious injuries,
which can be significant.

The study would have provided a valuable assessment of the benefits
to risk of the Bicing system had it taking more realistic assumptions of
the level to which the system was replacing motor vehicle use.

In the concluding comments, the authors note that the Bicing scheme
has successfully increased the number of people cycling in Barcelona - to
a much greater degree than other initiatives in the past. An
unsubstantiated claim made by the authors suggests the Bicing scheme is
"low cost", yet no mention of Bicing's cost is used in the paper.
Similarly, the authors conclude that other cities should follow Barcelona
in creating "cost saving" transport initiatives - yet no benefit cost
analysis was undertaken in order to determine if the scheme produces a net
economic benefit.

Despite the assumptions compromising the findings of this study, the
authors made a number of insightful points that have not been widely
published in the literature on PBSS. Firstly, the authors highlight the
potential for bicycle trips to replace car trips of greater distance
(people choose closer destinations when they are solely under human
power). Secondly, Bicing and other PBSS might be the catalyst for
increasing the acceptability and legitimacy of cycling, which could act as
a catalyst for more cycling generally - even outside of the scheme. With
PBSS proving popular not just in continental Europe, but also in cities in
the UK and the US, it appears these innovative, public transport options
may help increase physical activity, reduce air and noise pollution and
reduce traffic congestions. Measures aimed at increasing the substitution
rates from car journeys will maximise these benefits.

Overall, the paper is a useful addition to the literature, in that it
has attempted to assess the health benefits of a large scale PBSS and
weighed these against potential risks related to cyclists exposure to air
pollution and road traffic injuries. Unfortunately a fundamentally flawed
assumption related to the proportion of Bicing trips replacing car
journeys invalidates the results of this paper. A future paper with up to
date data would create a significant contribution to this emerging area
within the field of sustainable transport.


Anaya, E., & Bea, M. (2009). Cost-benefit evaluation of Bicing.
Paper presented at the ECOMM Conference, San Sebastian/Donostia, Spain.

Buttner, J., Mlasowsky, H., Birkholz, T., Groper, D., Fernandez, A.
C., Emberger, G., . . . Banfi, M. (2011). Optimising Bike Sharing in
European Cities: A Handbook: Intelligent Energy Europe program (IEE).

City of Barcelona. (2007). The City Council extends Bicing services
Retrieved from

City of Barcelona. (2011). Bicing Press Release. Barcelona: City of

Fishman, E., Ker, I., Garrard, J., & Litman, T. (2011). Cost and
health benefit of active transport in Queensland. Brisbane: Produced for
Queensland Government.

New Zealand Transport Agency. (2009). Economic Evaluation Manual
(Volume 2) (pp. 276): New Zealand Transport Agency.

Rojas-Rueda, D., de Nazelle, A., Tainio, M., & Nieuwenhuijsen, M.
J. (2011). The health risks and benefits of cycling in urban environments
compared with car use: health impact assessment study. British Medical

Shaheen, S., Zhang, H., Martin, E., & Guzman, S. (2011). Hangzhou
public bicycle: Understanding early adoption and behavioural response to
bike sharing in Hangzhou, China. Paper presented at the TRB Annual
Meeting, Washington D.C.

Competing interests: No competing interests

18 August 2011
PhD Scholar
Centre for Accident Research and Road Safety - Queensland, Australia