Evaluation of the country random breath testing and publicity program inVictoria, 1993-1994

Monash University Accident Research Centre - Report #126 - 1997

Authors: M. Cameron, K. Diamantopoulou, N. Mullan, D. Dyte & S. Gantzer

Full report in .pdf format [6.5MB]

Abstract:

In November 1993, the Victoria Police in conjunction with the Transport Accident Commission launched a major program in country Victoria in an effort to increase the number of random breath tests (RBTs) to at least 700,000 tests in a twelve month period, and supported the enforcement operations with mass-media publicity. Monash University Accident Research Centre conducted an evaluation of the program with the main objectives being to evaluate the program in terms of implementation characteristics and its effects on road trauma.

From the introduction of the program in November 1993 to the end of 1994, 790,445 random breath tests were conducted in country Victoria. There was substantial variation in RBT activity patterns between the country Police Districts. The intensity of the supporting publicity was also measured, both directly and in terms of awareness levels in country areas. There was a statistically significant 9% reduction in HAH (high alcohol hour) serious casualty crashes below expected levels in country Victoria during the program.

Crashes in smaller areas of country Victoria influenced by the RBT activity were analysed so that the effects of styles of RBT operations, and the interactions of these effects with the levels of publicity awareness, could be seen. A statistically significant reduction in HAH serious casualty crashes was found when RBT was conducted by cars operating alone during the weeks and in the regions when enforcement was present. In addition, medium levels of awareness appeared to increase the effects of the "car only" enforcement operations. However, in regions and weeks influenced by car and bus combinations, a statistically significant net percentage increase in HAH serious casualty crashes occurred when high publicity awareness accompanied the enforcement.

These somewhat surprising findings led to the final stage of the evaluation which aimed to test the hypothesis that some drink-drivers faced with intense enforcement, heightened by intense publicity, changed their travel behaviour and used relatively unsafe minor roads, with negative consequences for road safety. The results of this component of the evaluation found that at times of intense enforcement and high publicity awareness in rural areas of Victoria, significant increases in HAH serious casualty crashes occurred on minor roads but not on major roads. These findings provide the basis for a decision to schedule RBT operations on minor roads as well as on major roads.

Executive Summary:

In November 1993, the Victoria Police in conjunction with the Transport Accident Commission launched a major program in country Victoria in an effort to increase the number of random breath tests (RBT) to at least 700,000 tests in a twelve month period, and supported the enforcement operations with mass-media publicity. The objectives of the research project were to evaluate the program in terms of implementation characteristics and its effects on road trauma, and to provide information to optimise the mix and levels of the major components of RBT programs (enforcement activities and supporting publicity) in the future.

From the introduction of the program in November 1993 to the end of 1994, 790,445 random breath tests were conducted in country Victoria, with the relatively high level of RBT activity achieved in Melbourne in 1993 being maintained during 1994. The Traffic Alcohol Section (TAS) booze buses appeared to display considerable deterrent value for drink-driving by random breath testing a greater number of drivers per hour than either District buses or cars across country Victoria. However, there was substantial variation in RBT activity patterns between the country Police Districts. The intensity of the supporting publicity was also measured, both directly and in terms of awareness levels in country areas.

GENERAL EFFECTS OF THE PROGRAM

A statistically significant 9% reduction in high alcohol hour (HAH) serious casualty crashes was found in country Victoria during the program, after taking into account other factors influencing trends in these crashes. This was greater than the non-significant reduction in crashes of the same type in Melbourne during the period, and can be contrasted with the (non-significant) increase in such crashes in country NSW when analysed in the same way. The extent to which it could be concluded that the reduction in HAH crashes in country Victoria may have been a general effect of the country RBT and publicity program was related to the findings regarding the localised effects of the enforcement operations in the regions and weeks where the RBT operated.

LOCALISED EFFECTS OF THE PROGRAM

The initial analysis of the localised effects on crashes of the country RBT and publicity program found evidence of reductions in HAH serious casualty crashes influenced by certain styles of RBT operations and/or when certain levels of publicity awareness accompanied the enforcement.

It should be noted, however, that the changes found in HAH crash frequencies do not represent an estimate of the effects of drink-driving advertising per se. They represent an estimate of the effect of each RBT style, conditional on the level of publicity awareness at the time the RBT activity was operating.

A statistically significant reduction in HAH serious casualty crashes was found when RBT was conducted by cars operating alone, during the weeks and in the regions when enforcement was present (estimated 22% reduction). There was also some evidence of an interaction between the effects of the enforcement operations and the levels of awareness of drink-driving television advertising in country Victoria. Medium levels of awareness appeared to increase the effects of the "car only" enforcement operations (estimated 33% reduction). There was little evidence of crash reductions during weeks when the RBT activity was accompanied by high levels of awareness of drink-driving publicity, or when combinations of cars and buses were operating together. In regions and weeks influenced by car and bus combinations, a statistically significant net percentage increase in HAH serious casualty crashes occurred when high publicity awareness accompanied the enforcement.

These somewhat surprising findings led to the final stage of the evaluation, which aimed to test whether the crash changes were consistent with the possibility that some drink-drivers faced with intense enforcement, heightened by intense publicity, changed their route and used relatively unsafe minor roads, with negative consequences for road safety.

LOCALISED EFFECTS ON CRASHES ON MAJOR AND MINOR ROADS

Evidence was found to support the hypothesis that during the 1993-1994 country RBT and publicity program, some drink-drivers faced with intense enforcement, heightened by intense publicity, changed their travel behaviour. The evidence was consistent with an increased use of relatively unsafe minor roads, resulting in increased serious casualty crashes in these circumstances, namely:

  1. RBT conducted by cars operating alone during the weeks when enforcement was present produced a statistically significant 24% reduction in HAH serious casualty crashes on major roads in the enforced regions. A 20% reduction also occurred for minor road HAH serious casualty crashes, however this reduction was not statistically significant and may have been due to chance.
  2. Cars and buses operating together during the weeks of enforcement were associated with an estimated 37% increase in HAH serious casualty crashes on minor roads in the enforced regions. The magnitude of the increase was similar for major road crashes. Neither increase, however, was statistically significant.
  3. There was some evidence of an interaction between the type of road on which serious casualty crashes occurred and the levels of awareness of drink-driving television advertising in country Victoria.
  • When low levels of publicity awareness were operating, a 29% net reduction occurred for HAH serious casualty crashes on minor roads in regions and weeks where any style of RBT was present. A smaller net reduction of 11% occurred during the weeks when publicity awareness was at a medium level. Neither of these reductions, however, were statistically significant.
  • Conversely, there was evidence of an increase in HAH serious casualty crashes on minor roads during weeks when the RBT activity was accompanied by high levels of publicity awareness. A statistically significant net increase of 81% resulted. In comparison, for HAH serious casualty crashes occurring on major roads a net decrease of 6.5% resulted; however this was not statistically significant.
  1. For minor road crashes, there was evidence of an interaction between the style of RBT operation and the levels of awareness of drink-driving television advertising in country Victoria.
  • When medium levels of publicity awareness were operating, a statistically significant 33% net reduction occurred for HAH serious casualty crashes on minor roads in regions and weeks where car-only operations were present.
  • When low levels of publicity awareness were operating, a 32% net reduction occurred for HAH serious casualty crashes on minor roads in regions and weeks when booze buses or a combination of cars and buses were present. However when medium levels were operating a 14% net increase resulted. Neither of these net changes in crash frequencies were statistically significant however.
  • Conversely, a statistically significant net increase of 311% in HAH serious casualty crashes occurred on minor roads in regions and weeks when buses or car/bus combinations were accompanied by high levels of publicity awareness. This increase is estimated to represent an additional 19 crashes occurring under these circumstances.

The above results suggest that at times of intense enforcement and high publicity awareness in rural areas of Victoria, significant increases in high alcohol hour serious casualty crashes occurred on minor roads but not on major roads. These findings provide the basis for a decision to schedule RBT operations on minor roads as well as on major roads in country areas.

This project was funded by the Transport Accident Commission