This is an adaptation of an approach that was used really successfully in the HIV space. Basically, you meet somebody's needs in terms of treatment, and you have positive knock-on effects in terms of the spread. In the case of HIV, you actually reduce the transmission of HIV among people if you provide them with medications like highly active antiretroviral therapies.
This is a slight adaptation of that approach, but of course drug use is a very, very different phenomenon. Essentially, we found in the work that I referred to in my opening remarks, funded by both NIDA in the United States and CIHR here in Canada, that people who were provided with opioid agonist treatments and who were injecting drugs were less likely to report that they had assisted in the initiation of other people into injection drug use. We know that injection drug use is often implicated in an increasing severity of opioid use disorder or other substance use disorders. We also found, for instance, that increasing the intensity of policing actually had the reverse effect. People who were encountering police more often were more likely to assist people in their initiation of injection drug use.
Let me just say that this is not to cast people who engage in this behaviour as predators or anything like that. There are many rational reasons that people engage in this kind of behaviour, but if we're looking to prevent the expansion of substance use behaviours that we think could potentially put people at higher risk of overdose and we rely on the evidence of interventions that can help meet people's needs themselves, we find that there may be this potential knock-on effect on other people being at risk.
On that I'll say that we have not seen the same evidence of the effectiveness of recovery-based treatment as opposed to opioid agonist treatment and pharmacotherapy treatment. I would point to a recent study—it will be coming out in Drug and Alcohol Dependence in January, but it's available online now—that compared overdose mortality among people who had been enrolled in methadone and buprenorphine with recovery-based non-pharmacotherapy treatments. It found that there was a reduced risk of overdose mortality among people who were enrolled in buprenorphine. However, when the authors looked at non-pharmacotherapy recovery-based treatment, there was an increased risk, compared with the placebo, of overdose mortality.
On that note, I would say that the adoption of the Alberta model, while it is of course aspirational.... I think everybody in this field who devotes their time to it is aspirational and optimistic about the possibilities of people becoming well, managing their lives, being healthy and having social well-being. In Alberta, after the adoption of the Alberta model in mid-2019, there was actually a more than doubling of the overdose mortality rate in that province. There was an increase in overdose mortality basically everywhere in Canada, but the rate of increase in Alberta actually outpaced a lot of other places in Canada, so I would just offer a little bit of caution on that.