IT is four years since Scotland lowered the legal drink drive limit from 80mg to 50mg of alcohol per 1000 ml of blood, bringing us into line with most of Europe and parts of Australia.

And yet the impact recorded in these nations has not been mirrored in Scotland.

Read more: Lancet study finds 'no significant change' in road crashes after Scotland lowers drink-drive limit

A previous European study of 15 countries found that an equivalent law change was associated with a 7.4% reduction in road fatalities.

In the Australian states of New South Wales and Queensland, where the threshold was cut in the 1980s, there was a 7% and 14% respectively in the number of severe road traffic accidents (RTAs), rising to 8% and 18% reductions for fatal RTAs.

For Scotland to record “no significant change” in its RTA rate in the two years after the law change, compared to the two years prior, is - to say the least - unexpected.

For England and Wales, where the drink-drive limit remains unchanged, to have experienced a reduction in RTAs of some 6% over the same period - making Scotland’s position appear even worse by comparison - is surely all the more galling.

Read more: Scotland to introduce drug-driving limits and testing

But the takeaway from today’s Lancet study is not that the policy is necessarily misplaced, but that it “was not suitably enforced - for example with random breath testing measures”.

In Australia, “most of the decrease in traffic-related injuries and fatalities has been attributed to the implementation of random breath tests”.

Initial investment in publicity campaigns after the launch period of December 2014 was also “not maintained in the subsequent years”.

“We know that for an intervention such as this to work, the public need to know about the legislation change - but there also needs to be a fear of being caught,” said Professor Jim Lewsey, one of the lead authors in today’s Lancet study.

A parallel study at Stirling University, due in 2019, is evaluating the policy through interviews with the public, police and alcohol retailers.

Prof Lewsey added: “The police interviews might give insight into whether police resources were being increasingly stretched.”

Read more: Revealed - Scotland's drink driving hotspots 

Reacting to the findings Alison Douglas, chief executive of Alcohol Focus Scotland, said the charity favours “introducing random breath testing” based on other countries experiences.

But as the Scottish Government stresses, key levers associated with drink-drive legislation - such as changes to penalties to deal with drink driving, or the ability to introduce ‘any time, anywhere’ breath-testing - are reserved to the UK Government.

Currently, random roadside breath-testing is restricted to specific campaigns, such as increased festive patrols.

Otherwise traffic police must suspect a motorist is over the limit - for example, driving erratically - before pulling them over.

For Neil Greig, the Scotland-based director of policy for charity IAM RoadSmart, the key issue is that the lower limit was never designed to deter the serial offenders.

“The people who are being caught now, are not being caught at the new lower limit - they’re being caught at or above the old limit,” said Mr Greig.

“We haven’t seen the ‘morning after’ problem that we thought we might get, although that could also be an enforcement issue, so the problem for us is that there seems to be a hardcore of people out there who are pretty much impervious to all this.

“No matter what you do with campaigns and education, it’s finding those people and having the police resources to catch them.

"The law-abiding continued to be law-abiding - maybe they stopped at one drink or didn’t drink at all - but it hasn’t affected the hardcore of drink drivers.”

One caveat raised by the authors is that we simply live “in an era of improved road safety...where drink driving is increasingly unacceptable”.

As a result, the more substantial reductions in RTAs seen historically “might be more difficult to obtain”, however much you reduce the drink-drive limit.