Distribution of rapid HCV antibody self-test kits via needle/syringe dispensing machines: Implementation and evaluation of the Vend-C pilot study in Melbourne, Australia.

Recent guidance from the World Health Organization strongly recommended hepatitis C virus (HCV) self-testing. We implemented the Vend-C pilot study to explore the effectiveness and feasibility of distributing rapid HCV antibody self-test kits to people who inject drugs via needle/syringe dispensing machines (SDMs). Over a 51-day study period between August and September 2022, we distributed HCV antibody self-test kits via two SDMs. During the study period, 63 self-test kits were dispensed, averaging 1.2 self-test kits per day. Our access methods for evaluation questionnaires failed to attract participants (n = 4). We implemented the Vend-C pilot study in direct response to recent WHO recommendations. While self-test kits were effectively distributed from the two SDMs, our evaluation methodology failed. Consequently, we cannot determine the success of linkage to care. Even so, with HCV treatment numbers dropping in Australia, innovative engagement solutions are needed, and considering the number of self-test kits provided in our pilot, the model could have an important future place in HCV elimination efforts.

'We don't live in a harm reduction world, we live in a prohibition world': tensions arising in the design of drug alerts.

BACKGROUND: Drug alerts designed for health and community workforces have potential to avert acute harms associated with unpredictable illicit drug markets, by preparing workers to respond to unusual drug-related events, and distribute information to service users. However, the design of such alerts is complicated by diverse needs of individuals, and broader socio-political contexts. Here, we discuss the tensions that arose in the process of co-designing drug alert templates with health and community workers. METHODS: We conducted five in-depth digital co-design workshops with 31 workers employed in alcohol and other drug and urgent care settings. Our approach to analysis was informed by Iterative Categorisation and reflexive thematic analysis methods. RESULTS: We identified five key tensions. First, there is a need to provide comprehensive information to meet the information needs of a diverse group of workers with varying knowledge levels, while also designing alerts to be clear, concise, and relevant to the work of individuals. Second, it is important that alerts do not create 'information overload'; however, it is also important that information should be available to those who want it. Third, alert design and dissemination must be perceived to be credible, to avoid 'alert scepticism'; however, credibility is challenging to develop in a broader context of criminalisation, stigmatisation, and sensationalism. Fourth, alerts must be carefully designed to achieve 'intended effects' and avoid unintended effects, while acknowledging that it is impossible to control all potential effects. Finally, while alerts may be intended for an audience of health and community workers, people who use drugs are the end-users and must be kept front of mind in the design process. CONCLUSIONS: The co-design process revealed complexities in designing drug alerts, particularly in the context of stigmatised illicit drug use, workforce diversity, and dissemination strategies. This study has highlighted the value of developing these important risk communication tools with their target audiences to ensure that they are relevant, useful, and impactful. The findings have informed the development of our drug alert prototypes and provide local context to complement existing best-practice risk-communications literature.

Co-designing drug alerts for health and community workers for an emerging early warning system in Victoria, Australia.

BACKGROUND: Alerts about changes in unregulated drug markets may be useful for supporting health and community workers to anticipate, prevent, and respond to unexpected adverse drug events. This study aimed to establish factors influencing the successful design and implementation of drug alerts for use in clinical and community service settings in Victoria, Australia. METHODS: An iterative mixed methods design was used to co-produce drug alert prototypes with practitioners and managers working across various alcohol and other drug services and emergency medicine settings. A quantitative needs-analysis survey (n = 184) informed five qualitative co-design workshops (n = 31). Alert prototypes were drafted based on findings and tested for utility and acceptability. Applicable constructs from the Consolidated Framework for Implementation Research helped to conceptualise factors that impact successful alert system design. RESULTS: Timely and reliable alerts about unexpected drug market changes were important to nearly all workers (98%) yet many reported insufficient access to this kind of information (64%). Workers considered themselves 'conduits' for information-sharing and valued alerts for increasing exposure to drug market intelligence; facilitating communication about potential threats and trends; and improving capacity for effective responding to drug-related harm. Alerts should be 'shareable' across a range of clinical and community settings and audiences. To maximise engagement and impact, alerts must command attention, be easily recognisable, be available on multiple platforms (electronic and printable formats) in varying levels of detail, and be disseminated via appropriate notification mechanisms to meet the needs of diverse stakeholder groups. Three drug alert prototypes (SMS prompt, summary flyer, and a detailed poster) were endorsed by workers as useful for supporting their work responding to unexpected drug-related harms. DISCUSSION: Alerts informed by coordinated early warning networks that offer close to real-time detection of unexpected substances can provide rapid, evidence-based drug market intelligence to inform preventive and responsive action to drug-related harm. The success of alert systems requires adequate planning and resourcing to support design, implementation, and evaluation, which includes consultation with all relevant audiences to understand how to maximise engagement with information, recommendations, and advice. Our findings about factors impacting successful alert design have utility to inform the development of local early warning systems.

Achieving meaningful participation of people who use drugs and their peer organizations in a strategic research partnership.

BACKGROUND: Peer-led programs with people who use drugs (PWUD) have been a key characteristic of the harm reduction in many countries, including their involvement in research. However, peer involvement in research is often limited to recruitment, consultation, and reporting back, rather than a genuine collaboration in the priority setting, design, and conduct of research. PWUD peer organizations face ongoing challenges to demonstrate the depth of their knowledge of current and emerging issues within drug-using networks and the value of their peer insights for effective research and policy. The identification of benefits, barriers, and enablers for meaningful participation of PWUD in research has often been limited to methodological rather than system level factors. METHODS: This paper draws on the experiences and findings of the What Works and Why (W3) Project, a 5-year collaborative study with peer organizations. The study drew on systems thinking methods to develop a framework to demonstrate the role of peer organizations within their community and policy systems. The study required peer staff and researchers to undertake the simultaneous role of drivers, participants, and analysts in the research. To identify the learnings in relation to meaningful participation of PWUD peer organizations in research, we drew together the insights and experiences of peer staff and researchers across the 5 years of the study RESULTS: The W3 Project provided insights into the nuances of community-engaged research practice and the ongoing benefits, barriers, and enablers to the meaningful participation of PWUD and their peer organizations. These included system-level barriers and enablers beyond individual research projects or methodology. The capacity of research and peer organizations to maintain meaningful peer participation in research can be restricted or enhanced by the systems in which they are embedded. CONCLUSIONS: Recognizing peer organizations as active participants and drivers within community and policy systems can help clarify their unique and critical role in research. Achieving meaningful collaboration with PWUD peer organizations requires looking beyond good practice methods to the system-level factors with attention to the system-level benefits, barriers, and enablers.

Time for Australia to increase take away doses in opioid agonist treatment.

Opioid Agonist Treatment is the cornerstone of minimising harms related to opioid use, however its uptake is limited by a tightly regulated and stigmatising treatment environment. The COVID-19 pandemic necessitated relaxation of some treatment restrictions, with global evidence pointing to more patient-centred care in this time. In light of local evidence to support the safety of increased access to takeaway doses and a precedent set by the Substance Abuse and Mental Health Administration, we recommend adoption of the Australian Interim Medication Assisted Treatment of Opioid Dependence guidance in Australia.

Parents who inject drugs: Demographics, care arrangements and correlates for child placement in out-of-home care.

INTRODUCTION: Children in families affected by substance use disorders are at high risk of being placed in out-of-home care (OOHC). We aimed to describe the characteristics of parents who inject drugs and identify correlates associated with child placement in OOHC. METHODS: We used baseline data from a community-based cohort of parents who inject drugs (SuperMIX) from Melbourne, Australia. Participants were recruited via convenience, respondent-driven and snowball sampling from April 2008 to November 2020, with follow-up until March 2021. To explore correlates associated with child placement to OOHC, we used multivariable logistic regression and assessed for potential interactions between gender and a range of relevant covariates. RESULTS: Of the 1067 participants, 611 (57%) reported being parents. Fifty-six percent of parents reported child protection involvement. Almost half (49%) had children in OOHC. Nearly half of the parents lived in unstable accommodation (44%) and many of them experienced moderate-severe levels of anxiety (48%) and depression (53%). Female or non-binary gender, identifying as Aboriginal or Torres Strait Islander, experiencing assault and having more children were associated with child removal to OOHC. Of the 563 participants who reported their own childhood care status, 135 (24%) reported they had been removed to OOHC. DISCUSSION AND CONCLUSIONS: We identified high rates of child placement in OOHC among parents who inject drugs. There is a need for targeted health and social services, that are gender and culturally responsive, in addition to systems-level interventions addressing social inequities, such as housing, to support parents to care for their children.

Stigma by association among alcohol and other drug and harm reduction workers: Implications for workplace outcomes.

INTRODUCTION: The negative attitudes people hold towards those who use alcohol or other drugs (AOD) can also affect the people who work with this community, leading to lowered productivity and wellbeing. The impact of this stigma by association in the AOD and harm reduction sector is particularly significant because workers may have lived experience of AOD use and identify strongly with their client group. This study aimed to examine how stigma by association among health workers in the AOD/harm reduction sector relates to workplace outcomes. A secondary aim was to explore how lived experience influences experiences of stigma by association. METHODS: The research used a cross-sectional survey design and data collection occurred in 2023. Australian AOD/harm reduction workers (n = 228) completed an online survey assessing stigma by association as well as various workplace outcomes measures. RESULTS: Participants who reported experiencing more stigma by association experienced poorer workplace wellbeing, higher burnout and greater intentions to leave the AOD/harm reduction field. Experiences of stigma by association were unrelated to job satisfaction. Additional analyses revealed that participants with lived experience reported higher levels of job satisfaction and lowered intentions to leave the sector, but findings of stigma by association and its impacts on workplace outcomes did not differ from those without lived experience. DISCUSSION AND CONCLUSIONS: Identifying staff experiences of stigma by association and developing support and advocacy mechanisms to address this is likely to be key to reducing these experiences and ultimately to increasing positive workplace outcomes for AOD and harm reduction staff.

High COVID-19 vaccine uptake following initial hesitancy among people in Australia who inject drugs.

BACKGROUND: Previous studies have reported high COVID-19 vaccine hesitancy among people who inject drugs. We aimed to examine COVID-19 vaccine coverage, motivations and barriers to vaccination, and factors associated with uptake among this population in Australia, 1.5 years after vaccine rollout commenced. METHODS: In June-July 2022, 868 people (66.0 % male, mean age 45.6 years) who regularly inject drugs and reside in an Australian capital city reported the number of COVID-19 vaccine doses they had received and their primary motivation (if vaccinated) or barrier (if unvaccinated) to receive the vaccine. We compared vaccine uptake to Australian population estimates and used logistic regression to identify factors associated with ≥ 2 dose and ≥ 3 dose uptake. RESULTS: Overall, 84.1 % (n = 730) had received at least one COVID-19 vaccine dose, 79.6 % (n = 691) had received ≥ 2 doses, and 46.1 % (n = 400) had received ≥ 3 doses. Participants were less likely to be vaccinated than the Australian general population (prevalence ratio: 0.82, 95 % confidence interval [CI]: 0.76-0.88). Key motivations to receive the vaccine were to protect oneself or others from COVID-19, while barriers pertained to vaccine or government distrust. Opioid agonist treatment (adjusted odds ratio [aOR]: 2.49, 95 % CI: 1.44-4.42), current seasonal influenza vaccine uptake (aOR: 6.76, 95 % CI: 3.18-16.75), and stable housing (aOR: 1.58, 95 % CI: 1.02-2.80) were associated with receipt of at least two vaccine doses. Participants aged ≥ 40 years (versus < 40 years; aOR: 1.66, 95 % CI: 1.10-2.53) or who reported a chronic health condition (aOR: 1.71, 95 % CI: 1.18-2.47) had higher odds of receiving at least three vaccine doses. CONCLUSION: We observed higher COVID-19 vaccine uptake than expected given previous studies of vaccine acceptability among people who inject drugs. However, it was lower than the general population. People who inject drugs and reside in unstable housing are a subpopulation that require support to increase vaccine uptake.

Peer support models for people with a history of injecting drug use undertaking assessment and treatment for hepatitis C virus infection.

People who inject drugs (PWID) are the group most affected by HCV; however, treatment uptake has been low. Engagement between PWID and healthcare workers has been characterized by mistrust and discrimination. Peer support for HCV is one way to overcome these barriers. Peer support models for chronic disease management have been successfully applied for other diseases. HCV peer support models have been implemented in various settings, but those that include opioid substitution treatment have been more common. Most models have been either service generated (provider led) or community controlled (peer led). Peer support models have been implemented successfully, with a range of outcomes including increased treatment knowledge and uptake and improved service provision. Genuine partnerships between peers and services were common across models and led to positive transformations for both clients and services. Further investigation of peer support for HCV treatment and its impact on both individuals and services is recommended.

People who use drugs in hospital settings: 'Moving towards a person-centred harm reduction model'.

People who use drugs often continue to use drugs while in hospital. However, health-care systems often expect abstinence from drugs as a condition of engagement in various services. This commentary piece proposes that this approach is inconsistent with the principles of person-centred care. A harm reduction-based approach in conjunction with collaboration of people who use drugs is proposed as a model for providing person-centred care to people who use drugs during hospital-based treatment.

People engaged in opioid agonist treatment as a counterpublic during the COVID-19 pandemic in Australia: A qualitative study.

INTRODUCTION: People receiving opioid agonist treatment (OAT) are at higher risk of comorbidities, poverty and discrimination, which Big Events like the COVID-19 pandemic may exacerbate. The behaviours of people receiving OAT do not always align with normative behaviours as conceived by ruling institutions and laws, and so the group becomes a counterpublic, not imagined in mainstream public discourse. The aim of this study was to understand how people receiving OAT, as a counterpublic, implemented practises of care to mitigate negative health outcomes during COVID-19. METHODS: Participants were recruited via eight peer-led organisations across Australia. In-depth, semi-structured interviews were completed between August and December 2020 with 40 people receiving OAT. The analysis centres practises of care, allowing interactions that influence the health of participants, to be understood in their unique contexts. RESULTS: Aspects of the COVID-19 state response were designed for an idealised public, demonstrated by the increased policing that accompanied enforcement of restrictions which was detrimental to the wellbeing of people receiving OAT. Counterpublic health strategies employed by people receiving OAT were disrupted, but participants were often able to adapt to the changing context. DISCUSSION AND CONCLUSION: This study elucidates how practises of care among people receiving OAT are enacted and disrupted during a Big Event, with implications beyond the COVID-19 pandemic for future Big Events. The study findings evidence the need for policies that mitigate the impact of Big Events such as supporting re-groupment within the counterpublic, legitimising counterpublic health strategies and stopping the criminalisation of people who use drugs.

Information acquisition and dissemination among a sample of people who inject drugs in Australia.

INTRODUCTION: This paper examines the acquisition and dissemination of harm reduction information among people who inject drugs, as well as preferred sources of information. METHODS: Data were obtained from 862 people who inject drugs, recruited from Australian capital cities for the 2021 Illicit Drug Reporting System. Multivariable regression analyses were performed to assess potential factors associated with knowledge sharing. RESULTS: Almost two-fifths (37%) reported that they had received information about how to keep themselves safe when using drugs within the past 6 months. Reporting on their last occasion of receiving information, participants stated that it was commonly about injecting practices (56%), overdose prevention (26%) and injection-related injuries (22%), and was mostly received from an alcohol and other drug worker (54%), followed by other health professional (24%) and social network (18%). Among those who reported receiving information, 50% shared this information with other people, predominantly with their social network: no factors were found to be significantly associated with sharing information. The majority reported that peer workers and/or people with lived experience would be the first person they would talk to for information about a range of topics (e.g., injecting/harm reduction practices, overdose prevention). DISCUSSION AND CONCLUSIONS: Two in five participants had recently obtained information about how to keep themselves safe while using drugs, with half sharing this information with their social network. Peer workers were the preferred source of information, suggesting that the peer educator workforce should be expanded to embrace the capacities and expertise of people who inject drugs.

Deimplementation in the provision of opioid agonist treatment to achieve equity of care for people engaged in treatment: a qualitative study.

BACKGROUND: Deimplementation, the removal or reduction of potentially hazardous approaches to care, is key to progressing social equity in health. While the benefits of opioid agonist treatment (OAT) are well-evidenced, wide variability in the provision of treatment attenuates positive outcomes. During the COVID-19 pandemic, OAT services deimplemented aspects of provision which had long been central to treatment in Australia; supervised dosing, urine drug screening, and frequent in-person attendance for review. This analysis explored how providers considered social inequity in health of patients in the deimplementation of restrictive OAT provision during the COVID-19 pandemic. METHODS: Between August and December 2020, semi-structured interviews were conducted with 29 OAT providers in Australia. Codes relating to the social determinants of client retention in OAT were clustered according to how providers considered deimplementation in relation to social inequities. Normalisation Process Theory was then used to analyse the clusters in relation to how providers understood their work during the COVID-19 pandemic as responding to systemic issues that condition OAT access. RESULTS: We explored four overarching themes based on constructs from Normalisation Process Theory: adaptive execution, cognitive participation, normative restructuring, and sustainment. Accounts of adaptive execution demonstrated tensions between providers' conceptions of equity and patient autonomy. Cognitive participation and normative restructuring were integral to the workability of rapid and drastic changes within the OAT services. Key transformative actors included communities of practice and "thought leaders" who had long supported deimplementation for more humane care. At this early stage of the pandemic, providers had already begun to consider how this period could inform sustainment of deimplementation. When considering a future, post-pandemic period, several providers expressed discomfort at operating with "evidence-enough" and called for narrowly defined types of data on adverse events (e.g. overdose) and expert consensus on takeaway doses. CONCLUSIONS: The possibilities for achieving social equity in health are limited by the divergent treatment goals of providers and people receiving OAT. Sustained and equitable deimplementation of obtrusive aspects of OAT provision require co-created treatment goals, patient-centred monitoring and evaluation, and access to a supportive community of practice for providers.

'I just thought that was the best thing for me to do at this point': Exploring patient experiences with depot buprenorphine and their motivations to discontinue.

INTRODUCTION: Long-acting injectable depot buprenorphine is a recent addition to the suite of opioid agonist therapies (OAT) used to treat opioid use disorder (OUD). However, there has been little research that focuses on the lived experience of people receiving depot buprenorphine treatment and reasons for why people decide to discontinue. The aim of this study was to explore what it is like to receive depot buprenorphine and to understand the motivations behind why people discontinue. METHODS: Open-ended, semi-structured interviews were conducted between November 2021 and January 2022 with individuals who were either currently receiving depot buprenorphine or had discontinued or were in the process of discontinuing depot buprenorphine. Liberati, et al.'s (2022) adaptation of Dixon-Woods's (2006) candidacy framework was used to analyse the participant experiences. RESULTS: 40 participants (26 male, 13 female, 1 undisclosed; mean age 42 years) were interviewed about their experience with depot buprenorphine. At the time of the interview, 21 were currently receiving depot buprenorphine and 19 had discontinued this treatment or were in the process of discontinuing. Participants cited 4 key reasons why they decided to discontinue depot buprenorphine:1) feeling forced into the program, 2) experiencing negative side-effects, 3) finding the treatment ineffective, and 4) wanting to stop depot buprenorphine/OAT to use opioids again or feeling 'cured' and no longer in need of OAT. Participants were ultimately discussing issues related to clinician-patient power relations, agency and bodily autonomy, and the pursuit of well-being. CONCLUSION: Depot buprenorphine remains a promising treatment for OUD and offers potential to improve treatment adherence. Instances of restricted OAT choice and consumer concerns regarding a lack of agency must be addressed in order to enhance therapeutic relationships. Clinicians and other healthcare workers in this field also need greater access to information about depot buprenorphine to better address issues patients face during treatment. More research is required to understand patient and treatment choice given the options of these new treatment formulations.

"You'll come in and dose even in a global pandemic": A qualitative study of adaptive opioid agonist treatment provision during the COVID-19 pandemic.

BACKGROUND: Opioid agonist treatment (OAT) improves multiple health and social outcomes, yet requirements to attend for supervised dosing can be burdensome and stigmatising. The COVID-19 pandemic and associated restrictions threatened continuity of care and the wellbeing of people receiving OAT, risking a parallel health crisis. This study sought to understand how adaptations in the complex system of OAT provision impacted and responded to risk environments of people receiving OAT during the COVID-19 pandemic. METHODS: The analysis draws on semi-structured interviews with 40 people receiving and 29 people providing OAT located across Australia. The study considered the risk environments that produce COVID-19 transmission, treatment (non-)adherence, and adverse events for people receiving OAT. Drawing on theories of risk environments and complex adaptive systems, data were coded and analysed to understand how adaptations to the typically rigid system of OAT provision impacted and responded to risk environments during the COVID-19 pandemic. RESULTS: During COVID-19, the complex system of OAT provision demonstrated possibilities for responsive adaptation to the entangled features of risk environments of people receiving OAT. Structural stigma was evident in the services which stayed rigid during the pandemic, requiring people to attend for daily supervised dosing and risking fracturing therapeutic relationships. In parallel, there were several examples of services developing enabling environments by offering flexible care through increased takeaways, treatment subsidies, and home delivery. CONCLUSIONS: Rigidity in the delivery of OAT has been an impediment to achieving health and wellbeing over past decades. To sustain health-promoting environments for people receiving OAT, the wider impacts of the complex system should be acknowledged beyond narrowly defined outcomes relating solely to the medication. Centring people receiving OAT in their own care plans will ensure adaptations in the complex system of OAT provision are responsive to the individual's risk environment.

COVID-19 vaccine attitudes and facilitators among people in Australia who inject drugs.

INTRODUCTION: We aimed to describe COVID-19 vaccination attitudes and identify potential facilitators for vaccine uptake among people who inject drugs. METHODS: People who inject drugs were recruited from all eight Australian capital cities (N = 884; 65% male, mean age 44 years) and interviewed face-to-face or via telephone in June-July 2021. COVID-19 and broader vaccination attitudes were used to model latent classes. Correlates of class membership were assessed through multinomial logistic regression. Probability of endorsing potential vaccination facilitators were reported by class. RESULTS: Three classes of participants were identified: 'vaccine acceptant' (39%), 'vaccine hesitant' (34%) and 'vaccine resistant' (27%). Those in the hesitant and resistant groups were younger, more likely to be unstably housed and less likely to have received the current season influenza vaccine than the acceptant group. In addition, hesitant participants were less likely to report a chronic medical condition than acceptant participants. Compared to vaccine acceptant and hesitant participants, vaccine-resistant participants were more likely to predominantly inject methamphetamine and to inject drugs more frequently in the past month. Both vaccine-hesitant and resistant participants endorsed financial incentives for vaccination and hesitant participants also endorsed facilitators related to vaccine trust. DISCUSSION AND CONCLUSION: People who inject drugs who are unstably housed or predominantly inject methamphetamine are subgroups that require targeted interventions to increase COVID-19 vaccination uptake. Vaccine-hesitant people may benefit from interventions that build trust in vaccine safety and utility. Financial incentives may improve vaccine uptake among both hesitant and resistant people.

The impacts of COVID-19 measures on drug markets and drug use among a cohort of people who use methamphetamine in Victoria, Australia.

BACKGROUND AND AIMS: Few studies of the impacts of the coronavirus disease 2019 (COVID-19) public health measures on drug markets and drug use patterns have used longitudinal data. We aimed to examine whether COVID-19 measures were associated with increases in methamphetamine price, decreases in methamphetamine use frequency and subsequent changes in secondary outcomes of other drug use frequency in metropolitan Melbourne and regional Victoria. DESIGN: Longitudinal analysis framework was used from a longitudinal cohort of people who use methamphetamine. SETTING: Victoria state, Australia. PARTICIPANTS: One hundred eighty-five VMAX study participants who reported a methamphetamine purchase after the onset of the pandemic were used for the price paid analysis. Methamphetamine or other drug use frequency analysis was performed using 277 participants who used methamphetamine during the pandemic or in the year before the pandemic. MEASUREMENTS: Price paid per gram of methamphetamine derived from the most recent purchase price and most recent purchase quantity. Frequency of methamphetamine and other drug use measured as the average number of days per week used in the last month. FINDINGS: Compared with pre-COVID-19 period, methamphetamine prices increased by AUD351.63 (P value <0.001) and by AUD456.51 (P value <0.001) in Melbourne and regional Victoria, respectively, during the period in which the most intense public health measures were implemented in Victoria. Although prices decreased after harder restrictions were lifted (by AUD232.84, P value <0.001 and AUD263.68, P value <0.001, in Melbourne and regional Victoria, respectively), they remained higher than pre-COVID-19 levels. A complementary 76% decrease was observed in relation to methamphetamine use frequency in regional Victoria (P value = 0.006) that was not offset by any changes in the frequency of use of other drugs such as alcohol, tobacco or other illicit drugs. CONCLUSION: COVID-19 public health measures in Victoria state, Australia, appear to have been associated with major price changes in the methamphetamine market and decreased frequency of use of the drug.