Pilot study of promotion of public breath testing

Monash University Accident Research Centre Report No. 117 - 1997

Authors: N. Haworth, L. Bowland, P. Vulcan & C. Finch

Full report in .pdf format [311KB]

Abstract:

This pilot study evaluated the effectiveness of various promotional activities in attempting to achieve a 10% or greater usage rate of public breath test machines by drivers with a BAC over .05 in licensed premises in Melbourne. The two levels of promotional activity undertaken were placement of posters and coasters in venues (Level A) and posters and coasters plus an in-house promotional event and local media coverage (Level B). The cost (to the driver) of testing was 20 cents, $1 or $2.

The results of the study suggest that reducing the cost of testing to 20 cents would be a more effective way of increasing the number of drivers with BAC>.05 who test themselves, rather than committing resources to promotion. The greatest benefit cost ratios are obtained by using twenty cent machines: the estimated BCR for installing 300 twenty cent machines in the 150 highest volume liquor establishments is 1.41 if the net machine cost per month is $120. All other combinations of cost of testing and net machine cost per month lead to BCRs which are not statistically greater than 1.00. In addition, there may be some smaller volume establishments with higher than average usage rates for which installation of twenty cent machines might result in a BCR significantly greater than one.

To maximise the number of tests and thus the benefits in terms of road safety and commercial viability, there is a need to carefully target particular types of venues and locations for installation of BTMs.

Executive Summary

This pilot study evaluated the effectiveness of various promotional activities in attempting to achieve a 10% or greater usage rate of public breath test machines by drivers with a BAC over .05 in licensed premises in Melbourne. This information is needed to decide whether to commit substantial resources to promote the wider use of these machines.

The two levels of promotional activity undertaken were placement of posters and coasters in venues (Level A) and posters and coasters plus an in-house promotional event and local media coverage (Level B). The cost (to the driver) of testing was 20 cents, $1 or $2.

The usage rates and benefit-cost ratios were calculated for use by drivers only, rather than all patrons. The interview data showed that fewer than 50% of patrons (and users) were drivers. This reduced the usage rates and benefit-cost ratios markedly.

Objective 1: To determine whether any promotional activity or combination of activities will result in 10% or more of drivers with BACs>.05 (the target group) using breath testing machines (BTMs).

When all the experimental venues were included, the mean usage rate was significantly less than 10% both before and after promotions. The overall usage rate is dependent upon the number of machines at each cost of testing, however.

When analysed according to cost of testing, usage rates after Level A promotions (but not before) were significantly higher than 10% for venues with 20c machines. Usage rates were not significantly larger than 10% before or after promotion, for the $1 or $2 machines. While these analyses had sufficient statistical power, the results should be viewed with caution because it is based on very small numbers of venues and there was considerable variability between venues. If another sample of venues were chosen, the results may have been somewhat different.

Objective 2: To compare the relative effectiveness of different promotional activities in achieving increased usage rates for the target group.

Both levels of promotion were found to produce statistically significant increases in usage rates from the pre-promotion level.

However, analysis showed that the mean number of tests with BAC>.05 did not increase with promotion when controlled for external influences. The results suggest that the observed increases in usage rates resulted from a reduction in the size of the target group (the denominator of usage rate), rather than an increase in the number of tests by members of the target group (the numerator). The reduction in the size of the target group which was observed may have resulted from concurrent increases in drink driving publicity and enforcement.

Objective 3: To assess the commercial viability of breath test machines charging 20 cents, $1 or $2.

While usage rates of greater than 10% by the target group were achieved for 20 cent machines, the absolute number of tests, and thus revenue was not great. The costs at which machines would be financially viable, at each cost of testing, are presented in the report.

The results of the study suggest that reducing the cost of testing to 20 cents would be a more effective way of increasing the number of drivers with BAC>.05 who test themselves, rather than committing resources to promotion. The greatest benefit cost ratios are obtained by using twenty cent machines: the estimated BCR for installing 300 twenty cent machines in the 150 highest volume liquor establishments is 1.41 if the net machine cost per month is $120. All other combinations of cost of testing and net machine cost per month lead to BCRs which are not statistically greater than 1.00. In addition, there may be some smaller volume establishments with higher than average usage rates for which installation of twenty cent machines might result in a BCR significantly greater than one.

To maximise the number of tests and thus the benefits in terms of road safety and commercial viability, there is a need to carefully target particular types of venues and locations for installation of BTMs. Venues varied in the number of tests, the proportion of drivers who had BAC>.05 and usage rates. Size of the venue clearly constrains the number of tests, but other factors also play a role. Lower usage rates (not numbers of tests) at some larger venues suggest that perhaps more than one machine could be placed there. Placement of the machine in the venue can influence the number of tests: machines placed where people are drinking most (or most people are drinking) are likely to be used more frequently than those placed in, for example, gaming rooms where little alcohol is consumed. A larger proportion of drivers have BAC>.05 at some venues than others. Identification and targeting of these venues would assist in maximising the usage rates and road safety benefits of BTMs.

Sponsor: This research was undertaken under contract to VicRoads with financial support from the Transport Accident Commission and the Federal Office of Road Safety