This large observational study, containing 2736 patients receiving MAT for OUD, addressed the following: (1) differences in sociodemographic and clinical characteristics of men and women in treatment; (2) the SF of men and women in treatment; (3) if treatment is associated with improved SF; and (4) characteristics associated with SF.
Our results add to previous data showing that women are susceptible to a heightened burden of disease from OUD [41], with more women than men reporting unemployment, interpersonal conflict, as well as psychological and physical symptoms. The unemployment rate for women in our study was 74%, which significantly eclipsed the rate for men; both of which stagger above the rate of the Canadian general population (5% in 2018) [42]. Unemployment is a known social determinant of health; associations between unemployment and poorer health outcomes are well studied, including greater psychological distress [43], food insecurity [44], and possibly, suicide [45]. A systematic review and meta-analysis of methadone studies, many of which were conducted over 20 years ago, showed that women with OUD were at greater risk of unemployment, compared with men [27]. Our results show that the vocational landscape has not improved for women in MAT over the last two decades. A gap seems to exist in the social needs of patients in MAT; the need for integrated social and healthcare models of delivery for this large sector of the population, who are relatively young (in their 30s), is evident. Clearly, we must strive to better understand the substantial unemployment rate of MAT users, in particularly in women, through research.
Women in MAT were also more likely than men to report serious interpersonal conflict. As mentioned, interpersonal conflict was defined as verbal abuse, violence, or a major argument. That 46% of women and 35% of men endorsed conflict is alarming and likely to impact personal safety and quality of life, as well as impede recovery. Situating these findings in the pre-existing literature is challenging, given that no other studies have specifically assessed interpersonal conflict using the MAP. However, literature supporting associations, potentially causal and bidirectional in nature, between domestic abuse/violence and mental health difficulties is copious, and may be extrapolatable to our study population [46]. A review and meta-analysis found that women who had ever experienced partner violence were twice as likely to report depressive disorders, four times as likely to meet criteria for an anxiety disorder, and seven times as likely to be diagnosed with post-traumatic stress disorder (PTSD) [47]. This relationship is multidirectional and complex; as mental disorders predispose to substance use [48], substance use is a risk factor for partner violence [46], partner violence increases the risk for mental disorders [47], and mental disorders increase the perpetuation of partner violence [49]. Indeed, our study showed psychological symptoms increased the likelihood of interpersonal conflict in both men and women. There is clearly a need for future research into interventions to specifically address the perpetuation and experience of violence for people with mental health and addiction difficulties.
We found that men were more likely to report criminal activity, and also to report more episodes of crime per month, than women; results that confirm findings from older, small-sampled trials in the OUD population [27] and are consistent with statistics from the general Canadian public. Just 21% of Criminal Code offences are perpetuated by women and almost 94% of offenders in provincial custody have been identified as having a substance use problem [50]; the nature of this relationship remains murky and may be more pronounced for women [51]. Pierce et al.’s large cohort study revealed that having a history of any opioid use increased the likelihood of offending by two for men and by four for women. Compared with non-users, male and female opioid users had higher rates of criminal activity prior to opioid initiation, however, the relative risk strengthened post-opioid initiation [51]. These results are, however, limited by lack of control for important confounders, such as childhood adversity [52] and antisocial personality disorder [53].
Surprisingly, in our study, women were more likely to report criminal activity the longer they were on treatment (Table 3); every additional year on MAT was associated with a 7% increase in odds of reporting crime. Dealing drugs was the crime most often reported by women (data available upon request). This finding, coupled with the unemployment rate that remains dire for women throughout MAT, leads to one possible explanation. Women might increasingly deal drugs in attempt to mitigate persistent financial instability; however, we are unable to explore this hypothesis empirically. Results from Pierce et al. [51] highlight the potential for opioid-use prevention strategies for women in mitigating criminal activity; minimizing inappropriate prescribing and expanding purposeful work opportunities might be important approaches [54]. More work is needed to expand, and research prevention strategies for OUD, as well as better understand the role of MAT and OUD in offending, especially in women.
These results are at odds with a prior systematic review and meta-analysis of methadone studies based in China [55]. The included papers were deemed of low-quality and enrolled predominantly male cohorts. While it is difficult to draw comparisons to our study, the aforementioned meta-analysis did importantly show that longer duration of methadone was associated with reduced drug-related arrests and subtle reductions in self-reported drug-selling and sex-selling behavior.
Given the cross-sectional design of our study, the findings of increased criminality among women in MAT should be interpreted with caution. Theoretically, our findings could be explained by increased MAT retention in women with more criminal involvement. While the literature suggests the contrary, it is possible that in this case, more severe mental health problems and therefore criminal engagement may be linked to treatment seeking and study retention.
Nearly half of participants had initiated MAT within 1 year of study onset. As such, we conducted subgroup analyses to explore for differences in SF depending on treatment duration (i.e., less than, versus greater than, 1 year on MAT). Interestingly, compared to those relatively new to treatment, the group of men in treatment for more than 1 year reported less crime and conflict. It is therefore possible that men experience a delayed treatment response compared to women and that improvements in some aspects of SF may be gleaned only after 1 year.
After controlling for covariates, greater duration on MAT did not increase SF for men or for women. That being said, we did not collect data on most participants prior to treatment initiation. Thus, it is possible that participants’ SF might be substantially better than their pre-treatment state; a possibility unable to be answered by our data.
Indeed, participants in the previously mentioned NAOMI trial [21] reported a 24% increase in quality of life after 12 months of MAT. However, our study profiles may be different, as the NAOMI trial excluded some participants with severe medical and psychiatric comorbidities—confounders for both quality of life and SF. Participants in NOAMI were also treated with injectable diacetylmorphine and had psychosocial and primary care services available to them. Regardless, we present an important finding that deserves further exploration, as MAT is used as a harm reduction tool meant to improve several aspects of substance use disorder, not only the direct symptoms associated with drug effects.
Our study is the first to explore associations between SF and patient characteristics in MAT. Most noteworthy is that psychological symptoms increased the odds of poorer SF across domains (employment, criminality, conflict), for both men and women; however, the effect appears to be more profound for women. Taken in the context of the extant literature, these findings highlight the need to adequately manage distressful symptoms, particularly in women.
Continued use of illicit opioids was associated with criminal activity for both men and women in MAT. This finding affords credibility to a goal of abstinence for some people with OUD. As mentioned previously, selling drugs was the most common offence. Earlier studies have shown an association between selling and using drugs [56]. This relationship is likely bidirectional, as people might be inclined to sell drugs to support their drug use, and/or initiate their drug use because the supply is readily available; evidence from other fields certainly suggests a link between visibility and consumption [57]. It is also possible that opioids increase the risk of offending by exerting effects on cognition; the relationship between other modalities of intoxication and crime is well-documented [58].
Our study also elucidated key differences in men and women in MAT that might serve to inform future treatment programs. There is growing evidence to support tailoring addiction treatment programs to women in order to reduce potential barriers to treatment (e.g., child rearing duties); indeed, women’s only groups may have efficacy benefits [59].
Strengths and limitations
This study is strengthened by the multisite design and large cohort of participants. To our knowledge, this is the largest study to examine sex differences in MAT, as well as the SF of men and women receiving treatment for OUD. This is also the first study, to our knowledge, to explore the impact of MAT on multiple domains of SF. While addiction studies predominantly enroll men, the relatively equivalent numbers of men and women in our study, possibly representing a more clinically relevant sample [29], is an added strength.
In every comparison in the multivariate models, most confidence intervals were very near to one. Given the potential confounders that were not controlled for, such as duration of opioid use prior to treatment, there exists possibility that, in some cases, there were no significant associations between factors and SF domains.
As mentioned, length of time on MAT was not associated with greater SF for men or for women. As this study was cross-sectional, and we also had few participants just initiating MAT (the median treatment duration was 2 years at study entry; Table 1) and therefore, there is no way to compare SF with and without MAT. It is likely that SF is different for people who have never tried MAT. It thus remains conceivable that MAT does lead to improved SF in those with a worse starting point; this hypothesis should be explored through longitudinal studies examining the impact of MAT on SF outcomes in patients newly enrolled in MAT.
On the other hand, it could be that MAT alone is not enough to improve all domains of functioning for someone with an OUD. Indeed, a recent review of the literature yielded conflicting results; some MAT trials revealed no added benefit with counseling-based interventions, and others found improvements, particularly with contingency-management interventions [60]. It should be noted that the included studies only examined the effects of non-pharmacological intervention with MAT on opioid use and treatment retention, and thus, the impact on SF remains unknown in these studies. Given the already resource-constrained healthcare systems and the unabating opioid crisis, we must also strive to predict which MAT users are most in-need and will most benefit from additional psychosocial interventions. There is, however, a dearth of literature to guide us in the development of adjunctive interventions to address SF in MAT.
The utility of SF as an outcome in MAT research may be challenging to measure and may be impacted by several other factors. One reason is the discrepancy of what SF means and what measures best capture it. A study of SF in OUD [13] measured SF using The Social Functioning Scale, a measure designed to assess social adjustment in schizophrenia [61]. As mentioned previously, this study found worse SF in people on treatment for OUD, compared to matched controls. In order to improve comparability and consistency in the literature, a standardized measure of SF in OUD should be utilized, for example, the MAP [34] or similar measures with a great degree of overlap in order to combine individual studies for future systematic reviews and meta analyses to provide a conclusive summary of evidence.
A major limitation in MAT trials, and addictions research more generally, is the inconsistency in outcomes among studies [19, 62], as well as the reliance on poorly validated outcome scales [63]. Typically, the definition of “success” in addictions treatment is arbitrarily determined and likely due, in part, to convenience [63]. Most often the outcome of choice is abstinence [64] or treatment retention [28]. There is, however, greater recognition that success may be different depending on perspective; for instance, the policy-maker may stress mortality prevention, the healthcare provider may focus on abstinence, and the patient might value employment or having a better relationship with their partner. Reducing [64] or ceasing substance use [65], minimizing psychological distress [65], and re-engaging with daytime activities [65] were recovery goals deemed important by patients with other substance use disorders.
As is the case with all studies, our findings are limited by volunteer biases, as people with worse or better SF may have been less likely to participate, and those with an un-treated OUD may have inferior SF than those enrolled in the study. Given the observational study design, we are unable to draw causal conclusions from our findings.
Perspectives and significance
In summary, our results show that women in MAT experience worse SF than men. We also present evidence that our first-line treatment may not lead to sustained improvements in SF for people who are suffering. Indeed, women in MAT for longer reported greater engagement with criminal activity. While this study is not without limitations, we believe that the results raise important questions as to whether MAT alone can mitigate the dire SF that many people with OUD face.
