Using contextual information in the evaluation of the effectiveness of mid-platform fencing

Mid-platform fencing is one of the solutions that aims to minimise the number of rail suicides occurring at stations. This measure is in use on a number of stations across the UK, limiting public access to the main line tracks... [ view full abstract ]

Mid-platform fencing is one of the solutions that aims to minimise the number of rail suicides occurring at stations. This measure is in use on a number of stations across the UK, limiting public access to the main line tracks at station platforms where trains do not usually stop. To guide future approaches to prevention it is necessary to collect evidence of the effectiveness of this measure.

The evaluation of interventions for fatalities using only quantitative data can be difficult as these events are rare and traditional statistics are not suited to determining the significance of changes in numbers of events at a location. Furthermore, fitting a fence is not as simple as it sounds. It is important to take into consideration a wide range of factors, such as the type of the fencing, maintenance requirements or operation of gates in the fencing, the station design and typical movements and needs of passengers at the station. This paper describes how detailed, descriptive data have been used as part of an approach to evaluate the effectiveness of this type of fencing measure.

Preliminary data on the implementation of a fencing programme have been collected as part of the RESTRAIL project 2013-14. This included 52 stations, along three rail routes, 40 of which having either full or partial restriction to the main line using fencing. The current research study is revisiting the collection of data at a number of these locations to determine progress with the fencing programme and consider the effectiveness of the fencing initiative. This work includes visits to station, interviews with station and programme staff and review of statistical data on incidents arising at these stations.

Results will be provided for work at 20 locations where at least one fatality has occurred in the period after fitting of the fencing. These include collation of detail from the SMIS database and the BTP reports, producing descriptive data on the location of the incident (part of the station, line, point of access to the track), the level and type of restrictions in place at the time of the incident and other related factors. For example, evidence has been collected to explain how one person was able to climb over the fencing using poorly located station furniture. Interviews with industry staff and visits to stations have provided information on station characteristics (staffing levels, procedures for the control and operation of the fencing, other suicide prevention programmes at the location). These have been compared across the three rail routes in the study. The evidence shows that the provision of fencing does not guarantee the intended level of restriction of access to the main line. Examples are discussed where access has not been restricted in the way that it was intended because of operational reasons. Suggestions are made to improve the effectiveness of the fencing and a number of potential drawbacks of the fencing measure are discussed (e.g. it might shift the problem to other locations). The value of the qualitative approach is discussed.