Design

We designed a mixed method observational study (Fig. 1). The ACT-ATAC project began in January 2019 and was conceived as a 36-month study. However, due to the COVID-19 pandemic, we have extended the study for eight more months. At the present time, December 2021, we are in month 35 (Fig. 1) and we foresee the study lasting until the end of June 2023.

Fig. 1

The ACT-ATAC project flowchart

Mixed method studies include both a qualitative and a quantitative research perspective in which one of the study designs may have a lead role and the other has a more supporting role [32]. In this project, the cohort study is the primary design. The study is divided into the following substudies (S1-S3).

S1 (months 1–11): Phenomenological study with discussion groups directed towards health professionals (corresponding to January 2019–November 2019; carried out before the outbreak of COVID-19 in Spain).

This qualitative phenomenological study with discussion groups used a semi-structured script to answer aim 1. Our target was all healthcare professionals working at one of the 42 SUD programs in the province of Barcelona. Twelve centers were chosen from those that expressed interest according to a geographical distribution criterion to ensure variability of areas, as 65% of the centers in the province belong to the city of Barcelona and the rest to other nearby cities. Clinicians from these SUD centers were recruited via a letter signed by the Head of the Catalan Drug Plan and addressed to program directors, asking them to engage their staff in participating.

Participants

Participants included all healthcare professionals from the selected SUD programs who work in the mental health field with at least 1 year of experience in the field and agreed to participate and provide informed consent. Clinicians were grouped into three heterogeneous discussion groups (8 to 10 participants each) led by two experts in qualitative research who followed a script of themes to be explored. Themes included (1) barriers and opportunities to address tobacco use among persons in treatment for alcohol and/or cannabis use; (2) health professionals’ opinions on the type of smoking cessation interventions that could be offered to these patients and whether these interventions should be simultaneous or consecutive with the treatment of the main drug, and (3) experiences in helping smokers quit.

Procedures and statistical analysis

Sessions were recorded and transcribed, preserving participant anonymity and confidentiality, to conduct a thematic categorical content analysis (AC-CT). Thematic axes were proposed to finish constructing the questionnaire needed in the second phase of the study.

S2 (months 11–40): Longitudinal prospective study with smokers in treatment for alcohol and/or cannabis and the clinicians involved (corresponding to November 2019–December 2022; study piloted before the outbreak of COVID-19 in Spain but carried out during and after the severe lockdown of March 2020-June 2020, and currently ongoing).

The second substudy covers aims 2 and 3 and consists of two prospective cohorts composed of (1) smokers in treatment for alcohol and/or cannabis who visited one of the participating SUD programs, who agreed to participate in this study, and had at least one responsible healthcare professional to recruit participants during the first and second year of the study, and (2) the clinicians in charge of these patients.

Cohort of smokersParticipants

Inclusion criteria are (1) being a tobacco user (at least one cigarette per week), (2) initiating treatment for alcohol and/or cannabis in one of the SUD participating program, (3) age  ≥ 18 years, (4) under follow-up by participating clinicians, and (5) providing consent. Exclusion criteria are (1) inability to guarantee a 12-month follow-up (due to anticipated mobility problems, etc.) and (2) severe cognitive impairment and/or other communication problems (i.e., not speaking Spanish or Catalan).

Sampling method

The incidence of unassisted smoking cessation is seven ex-smokers per 100 patient-years [33]. We estimate that, among this sample, being in contact with the health system may increase this rate to 10 ex-smokers per 100 patient-years (i.e., a relative risk [RR] of quitting smoking of 2.0). Therefore, to detect this RR assuming an alpha error of 5%, a beta error  < 20%, and losses of 20%, we require 366 participants. The sample size was calculated using the GRANMO program [34]. This sample size is achievable in one year as the 48 centers in the province of Barcelona treat approximately 5000 new cases for alcohol and 1500 new cases for cannabis yearly (target population: 6,500 participants) [12]. On average, the 10–12 selected study centers should treat approximately 1,625 people/year. Smokers in treatment for alcohol and/or cannabis use are being recruited consecutively for one year.

Recruitment process and fieldwork

Clinicians working in the participating centers determine whether their patients meet the inclusion criteria described above and invite those patients to be part of the study. Those who accept to participate and complete informed consent procedures are registered using computer software designed for the ACT-ATAC project, including personal identifiers (i.e., name, surname), psychiatric diagnoses, comorbidities, and location data (contact telephone numbers, responsible clinician, clinician's telephone number, etc.). To facilitate the recruitment process, a protocol has been created in the software. After the participant’s data is introduced into the software, he/she is contacted in the first 48–72 h by a research nurse who conducts an extensive interview (see below). The study nurse is independent to any SUD participating center and was recruited exclusively to conduct this interviews. The study procedures are aimed at following up on the tobacco consumption of patients and other variables. The research nurse also informs the patient when he/she will be contacted for follow-up interviews by telephone (at 7 days and 3, 6, and 12 months; Fig. 2).

Fig. 2Fieldwork and strategies implemented to overcome the barriers due to the COVID-19 pandemic

The longitudinal fieldwork kicked-off in November 2019 (month 11). The first step of the fieldwork was recruiting SUD centers and working clinicians willing to participate in the project. The pilot study was conducted between November 2019 and February 2020 and allowed us to test the recruitment process and software designed to collect and store data, as well as ensure the correct follow-up of participants. Nevertheless, in March 2020, the COVID-19 pandemic in Spain forced us to modify the follow-up interviews from face-to-face to telephone-based to limit the social contact of the research nurse with the participants. In addition, the rhythm of patient inclusion in the study slowed down from 1 or 2 per week before the lockdown (from March to June 2020) to 2 or 3 per month. Despite this, the study was able to be continue because the SUD centers in Catalonia remained open, though many of the visits were conducted online. Thus, the collaborating clinicians were still able to recruit patients who met the inclusion criteria during the pandemic; however, we recommended that they did so during in-person meetings to better assess participants’ cognitive status and make getting their informed consent easier. These adjustments were possible thanks to the collaboration between the study coordinators, the collaborating clinicians, and the Direction of the Substance Abuse Program at the Public Health Agency in Catalonia, that prioritized our project and contacted clinicians to keep this study on track despite the contextual circumstances. To increase the engagement of clinicians (clinical psychologists, psychiatrists, and nurses), we have conducted monthly online meetings with them to exchange experiences and difficulties in the recruitment process and given them a certificate of participation in the study signed by the Director of the Substance Abuse Program at the Public Health Agency in Catalonia and the Principal Investigator. These meetings are an informal 30-min coffee break in which investigators and clinicians get together and share their experiences regarding the recruitment process and present the latest information on the participants’ profile. A mug with the main inclusion and exclusion criteria has been designed and gifted to clinicians for them to pick up during these meetings (Fig. 3).

Fig. 3

Clinician’s mug with the inclusion and exclusion criteria

Moreover, to improve participants’ fidelity in the follow-up study, halfway through the study, we started offering 20 euros gift cards to those who answer all of the phone calls during the 12-month follow-up. At the time of submitting this manuscript, 98 patients have been recruited. Thus, we will continue the study until we recruit at least 200 to 250 participants (expected by the end of 2022).

Variables

Dependent variables include tobacco use, cannabis use, and alcohol use. Tobacco use is rated yes/no. If answering “yes”, we ask the about the tobacco consumption pattern, including nicotine dependence using the 6-item Fagerstrom test [35], and the type of tobacco or other products consumed (i.e., cigarettes, roll-your-own tobacco, combination of both, cigars, e-cigarettes, iQOS); smoking history, including the age when the participant starting smoking, maximum abstinence time, previous quit attempts, and resources used (i.e., none, self-help, professional support, pharmacological treatment, etc.); motivation to quit smoking, measured by a Likert scale from 0 to 10; and willingness to quit smoking, assessed by a 4-item questionnaire that, according to Prochaska and DiClemente’s model, classifies smokers into five stages: pre-contemplative, contemplative, preparation, action, and maintenance [36].

Cannabis use is rated yes/no. If “yes”, we recorded users pattern of consumption, problematic use (yes/no) [37], withdrawal symptoms identified using the Marijuana Craving Questionnaire [38] and reasons for use [39]. Alcohol use is rated yes/no and described according to the variables of consumption included in the Alcohol Use Disorders Identification Test (AUDIT) [40, 41]. A fourth dependent variable is participants’ self-reported smoking knowledge, attitude, and services (S-KAS) received during their treatment in SUD programs [42].

The primary outcome is abstinence from smoking (yes/no), which is self-reported and verified by the exhaled CO concentration only if they report to be abstinent. Self-reported abstinence is defined as “no smoking in the 7 days prior to the assessment point.” Seven-day abstinence was chosen as the consumption measure based on the consensus recommendation in practice guidelines for the assessment of abstinence in patients who are not motivated to quit smoking and who may remain abstinent during various periods of the study [43]. Number of quit attempts in the last year of  > 24-h duration [44] were also recorded. All dependent variables are assessed at baseline and 3, 6, and 12 months of follow-up (Table 1).

Table 1 Study variables and collection time points (at baseline and follow-up)

Independent variables include sociodemographic data (sex, age, employment status and occupation, and education level), self-perceived state of health addressed by the question “How would you say your general health is?”, measures of mental and physical health according to the DSM-V [45], quality of life according to the General Health Questionnaire (GHQ-12) [46], and level of previous smoking cessation treatment received using questions about standard interventions included in the Smoking Cessation Guidelines (i.e., advice, reduction recommendation, educational material, cognitive-behavioral treatment, pharmacological treatment, nothing) [47].

Instrument

Baseline and follow-up questionnaires have been designed including all variables and their dimensions described above. These questionnaires are based on the findings of the discussion groups with clinicians and an extensive literature research. The backwards translation of the S-KAS [42] and the Marijuana Withdrawal Questionnaire from English to Spanish was performed by two Spanish native speakers. All data are entered into a computer software designed for the ACT-ATAC project for the introduction and management of patient data. The software allows recruitment and recording of initial variables by the clinician, and introduction of the results of the questionnaire interviews carried out by the field nurse. The initial and follow-up questionnaires (6 and 12 months) are not accessible by the clinician.

Cohort of cliniciansParticipants

Clinicians from the province of Barcelona who agree to participate in the study must meet the following inclusion criteria: (1) to be a clinician (psychologist, psychiatric, nurse, social worker, etc.) in one of the public SUD programs in the province of Barcelona, and (2) that does the initial assessment and follow-up of persons with alcohol and/or cannabis abuse disorders, and (3) provides consent. The exclusion criterion is an inability to guarantee a 12-month follow-up (due to anticipated moves, etc.). We expected to include between 20 and 30 clinicians, and 22 clinicians are currently collaborating in the study from 10 different SUD centers.

Variables

The primary dependent variable is self-reported smoking knowledge, attitude, and practice (S-KAP) [48] during the clinician’s work in the SUD program, monitored at baseline of their participation, one week after, and every three months during their participation in the study.

Independent variables are tobacco consumption (never user, former user, and current user), sociodemographic data (sex, age, education level, profession), and professional experience (years treating drug abuse population, years working in the same program). In the case of being a tobacco user, we ask about the tobacco product consumed (i.e., cigarettes, roll-your-own tobacco, combination of both, cigars, e-cigarettes, iQOS); smoking history, including age when starting smoking, maximum abstinence time, previous quitting attempts, and resources used (i.e., none, self-help, professional support, pharmacological treatment, etc.); motivation to quit smoking, measured by a Likert scale from 0 to 10; and willingness to quit smoking, assessed by a 4-item questionnaire according to Prochaska and DiClemente’s model, classifying smokers into five stages of change [36].

Clinicians receive the KAP questionnaire [48] every three months to report data on the approach taken in the care of the smoker (ask, advise, assist).