In a gender‐balanced sample of young adults in France who co‐use tobacco and cannabis, we found that material deprivation was directly associated with higher readiness to quit both substances and with reporting four or more cessation attempts for each substance. A lower SES was positively associated with readiness to quit cannabis when considering the total path (including direct and indirect effects). Lastly, we highlighted a direct association between higher cannabis use frequency and readiness to quit tobacco.

For both substances, our results suggest that financial strain fosters motivation to quit in young adults who co‐use. Saving money is indeed a commonly reported reason to quit tobacco [79, 80] and cannabis [71, 84] in young adults. Rising tobacco prices are an effective intervention to activate this lever to increase intention to quit and quit attempts, especially in low‐income groups [85, 86, 87]. However, price intervention alone is unlikely to translate in successful cessation in the most deprived individuals [86, 87]. For such populations, implementation of multicomponent interventions including elements such as social support and pharmacotherapy is critical [88].

While the published literature is inconclusive about the type of tobacco intervention that is most effective for young adults [89], we can expect, from our findings, that emphasising money savings when preparing a quit attempt or during it may foster motivation in young adults who co‐use. As a group, people who co‐use tobacco and cannabis are moreover likely to have lower financial resources [41].

This similar association of financial strain with readiness to quit and quit attempts for both substances suggests that financial incentives may be particularly suited to help people who co‐use to quit both simultaneously. Indeed, financial incentives have been shown to increase motivation to quit [90] and long‐term rates of tobacco smoking cessation [91, 92], especially among people with financial needs [93]. Few data are available for cannabis cessation [94, 95], but a recent randomised trial suggests that monetary incentive is effective at reducing cannabis use in people with heavy use [96]. A small pilot study also supported the feasibility and acceptability of monetary incentives in people who co‐use [97].

We found no overall association between low SES and readiness to quit tobacco, which seemingly reflects competing pathways. On the one hand, low SES is directly and positively associated with readiness to quit tobacco (supposedly through financial strain), and indirectly positively associated with it through cannabis use frequency. On the other hand, low SES is positively associated with tobacco use frequency, which is negatively associated with readiness to quit tobacco. The positive pathways appear to be cancelled out by the negative one. Similarly, frequency of use and financial strain showed opposing associations with the likelihood of reporting multiple tobacco quit attempts. A higher educational level was positively associated with having made at least 4 tobacco quit attempts, highlighting that all dimensions of the SES do not have similar effects on quitting motivation outcomes. In this small sample of young adults, our results therefore suggest that individuals with a high tobacco use frequency, a low educational level, but no strong financial strain, may be the least motivated to quit tobacco.

An original perspective that emerged from our cross‐sectional study is that a higher cannabis use frequency seemed to have a positive impact on readiness to quit tobacco. Some hypotheses can be suggested. First, we may expect some people with heavy cannabis use to endorse a cannabis user identity [98, 99, 100, 101], but wanting to get rid of the tobacco user identity which may be less desirable or more stigmatised [102, 103]. Given that cannabis may, to some extent, act as a substitute for tobacco [104, 105, 106, 107, 108], it is also plausible that, for some participants, readiness to quit tobacco led to a reduction in its use, which was subsequently compensated by an increase in cannabis consumption. Drawing on the compensatory health beliefs model [81], we may hypothesise that readiness to quit tobacco could serve as a ‘compensatory intention’, offsetting the perceived negative effects of frequent cannabis use. We may also hypothesise that some people using cannabis frequently employ tobacco solely as a mulling agent (e.g., to ensure proper combustion of cannabis joints [108]) and, as a result, may feel ready to quit tobacco—despite feeling technically constrained in their cannabis use. Finally, although no empirical evidence currently supports this hypothesis, individuals with limited financial resources may prioritise quitting tobacco to allocate more funds to cannabis, particularly if the latter is perceived as offering greater satisfaction, fewer side effects [108, 109], or reduced inconvenience compared to tobacco.

That said, it is important to consider the route of cannabis administration and what young adults mean by ‘tobacco cessation’. In France, as in Europe [40], cannabis is predominantly consumed in joints containing tobacco [110]. Thus, frequent cannabis use generally entails concomitant tobacco use through this route of administration. Quitting tobacco while continuing cannabis use would therefore require adopting a tobacco‐free route of cannabis administration (e.g., smoking pure cannabis, using a vaporizer or consuming edibles). It is also possible that by ‘tobacco cessation’, some participants understood ‘tobacco cigarette cessation’, perceiving their tobacco and cannabis use as disconnected [108], and thus considered quitting cigarettes while continuing to smoke tobacco‐containing joints. Such a situation would imply ongoing exposure to nicotine and could thereby foster relapse after tobacco cessation.

Qualitative studies are needed to shed light on those questions and hypotheses. The reverse association (i.e., a greater tobacco use frequency enhancing readiness to quit cannabis) did not reach statistical significance and should be further explored in studies with a larger sample size. Future findings in this direction might help understand potential superiority of sequential rather than simultaneous quitting in some people [111].

The associations we found suggest that reducing tobacco use may increase both readiness to quit tobacco and the likelihood of making a quit attempt. In contrast, they suggest that reducing cannabis use may decrease readiness to quit tobacco, while showing no significant effect on readiness to quit cannabis and unclear effects on the likelihood of attempting to quit. Those effects need to be confirmed by longitudinal studies, but our observations suggest that among people who co‐use, reducing cannabis use—a commonly observed behaviour [58]—may not constitute a desirable step towards dual cessation, whereas reducing tobacco use may be. It is worth noting that, according to a 2019 Cochrane review [112], neither reduction‐to‐quit nor abrupt tobacco cessation interventions showed superior long‐term quit rates compared with each other. To date, no clear superiority has been demonstrated between simultaneous and sequential dual cessation approaches [111].

In line with what was found elsewhere [113], cannabis dependence (vs. no cannabis dependence) showed a positive (albeit non‐significant) effect on readiness to quit cannabis. This discrepancy in the dependence‐to‐readiness relationships between tobacco and cannabis may partly reflect differences in the assessment tools used. The FTND primarily captures patterns of cigarette use and the inability to refrain from smoking [62] whereas the CUDIT‐SF items emphasise interference with daily functioning and negative consequences of cannabis use [63]. Further studies are needed to test how these two dimensions of substance‐use disorders influence motivation to quit tobacco and cannabis. Another possible explanation is the stronger intoxicating properties of cannabis: its acute and chronic effects on cognitive functioning [114] may encourage cessation because they impede daily functioning, while acute cognitive effects of nicotine tend to be beneficial [115].

In our sample, readiness to quit tobacco was higher than readiness to quit cannabis. This pattern is consistent with previous studies showing greater interest or desire to quit tobacco than cannabis among adults [106], young adults [116], college students [117] and adolescents [118] as well as with baseline findings from Becker et al.'s trial on dual cessation [119]. However, we observed somewhat more quit attempts for cannabis than for tobacco.

To explain this seemingly discrepancy, we may expect people who use cannabis to make many, and often rapid and unprepared, transitions between regular use, reduction and abstinence [58], resulting in a greater number of quit attempts according to our definition. Moreover, we may expect young adults to have lower expectancy of success with quitting tobacco than cannabis [116], resulting in less tobacco quit attempts. Lastly, as previously mentioned, cannabis‐related impairments in daily functioning may prompt spontaneous quit attempts, even without adequate preparation or elevated readiness to quit. By contrast, tobacco use is less likely to trigger such spontaneous quit attempts. Rather, its health‐related adverse effects are typically perceived only in the long term. For instance, older individuals are more likely to be motivated to quit tobacco by health concerns [120]. We can thus hypothesise that tobacco cessation attempts may require more mature forms of motivation, which are less likely to be present in young adults.

Our results are derived from cross‐sectional data, which precludes causal interpretation. Moreover, they rely on hypothesised directional associations that somehow remain questionable. For example, it is plausible that readiness to quit tobacco reduced tobacco use frequency, as tobacco use reduction may be an intermediate stage before cessation [121, 122]. The same may be true for cannabis. This uncertainty is further compounded by the possibility of model equivalence, whereby alternative causal structures cannot be distinguished based on our cross‐sectional data alone. Future longitudinal research is essential to clarify the directional nature of these relationships and validate our proposed model.

The first strength of this study lies in the assessment of different markers of motivation to quit both tobacco and cannabis. This approach makes it possible to examine the putative mechanisms underlying both motivations simultaneously among co‐users, and to compare them in parallel. Focusing on people who co‐use is original and enables us to extend some results found in studies exploring only one substance. Indeed, as tobacco use is frequent in people who use cannabis, especially in Europe [40, 41], we thus took into account the major cofounder, which is tobacco, when studying motivation to quit cannabis.

The models we used were expected to capture the direct and indirect effects of SES on readiness to quit. However, we acknowledge that factors unrelated to SES, which were not included in our models, may also impact readiness to quit, such as cognitive, psychological, or environmental factors. Our study population was gender‐balanced and came from diverse French regions. However, the recruitment process did not lead to a representative sample of young adults who co‐use because of recruitment biases and a slight overrepresentation of people in the Paris region (in 2025, it accounted for 21.6% of the 20–39 year population) [123]. Specifically, our recruitment procedure may have introduced a selection bias by favouring the inclusion of individuals already motivated to quit, while potentially excluding those less willing to disclose their substance use or motivations due to social desirability concerns. Consequently, our study sample may overrepresent individuals with higher quit intentions. Our method did not allow us to reliably assess the response rate. We cannot exclude the possibility that participants who completed the survey differed systematically from those who dropped out before completion, especially as we implemented forced answering. While there is no conclusive pattern of results for the impact of forced answering on data quality, its implementation may have impacted answer quality [124]. Additionally, the questionnaire did not include attention checks or internal reliability checks for responses.

Another limitation is the sample size, which was limiting in testing more complex pathways, such as treating material deprivation or financial strain as a separate latent variable (i.e., as a mediator between low SES and readiness to quit).

Due to the nature of our variables, we had to use GSEM, which, unlike SEM, does not allow for the calculation of goodness‐of‐fit indices such as the Comparative Fit Index or the Root Mean Square Error of Approximation. Readers should bear this limitation in mind when assessing the robustness of our conclusions. Cannabis use frequency was assessed in number of days of use per month, but joints per day, quantities, and cannabis potency were not taken into account, as well as possible other routes of administration, while we know they are important in characterising cannabis use and its effects [125, 126]. By limiting the administration of the FTND to people who smoke at least 25 cigarettes a day, we may have underestimated nicotine dependence. Lastly, tobacco use was assessed in cigarettes per day, but the role of tobacco within joints was not estimated.