Harnessing telemedicine to deliver patient-centered opioid agonist treatment within a community-based harm reduction service: a mixed methods evaluation

Abstract

Background

Patient-centered approaches may facilitate retention in opioid agonist treatment (OAT) but are challenging to implement in rigid or stigmatizing service contexts. We evaluate a telemedicine program delivering flexible, patient-centered OAT from a community-based harm reduction setting in Montreal, Canada.

Methods

An OAT clinic was established in a community-based harm reduction setting with hospital-based addiction medicine services delivered remotely via telemedicine. Community workers screened clients, established telemedicine connections, and offered holistic patient follow-up. The medical team offered individualized OAT and other health services. Patients chose between treatment with methadone, buprenorphine/naloxone, or slow-release oral morphine. Hydromorphone co-prescription was also available. Effectiveness was assessed via longitudinal chart review and semi-structured interviews (n = 20). A convergent mixed method design was used to quantify retention rates and blood borne infection care up to 12 month follow up, and to examine patient reported program experiences.

Results

Sixty-nine patients (46 men, 23 women; median age 38) initiated OAT between April 2020–March 2022. Most (96%) were injecting opioids, 56% were unstably housed, and 71% reported prior OAT. Patients typically initiated treatment with methadone (54%) or slow-release oral morphine (35%); 78% also received hydromorphone. Continuous retention in the first OAT episode was 83% at one month, 74% at three months, and 54% at 12 months. Disregarding prior treatment interruptions, 71% of patients were receiving OAT at 12 months. Most patients were assessed for HIV (77%) and HCV (78%), and 13/15 confirmed as HCV-positive initiated antiviral treatment. Trust, respect, and the alignment of practices with patient-centered care and harm reduction principles were critical to success. Integrating treatment within a community-based harm reduction setting enhanced accessibility and care coordination, and created a welcoming service environment. Diverse medication options, collaborative treatment planning, and a non-judgmental/non-punitive approach were key to developing positive therapeutic relationships.

Conclusion

Our community-based telemedicine program presents a novel framework for OAT delivery that efficiently bridges the health and community sectors. Working collaboratively around the patient, program partners leveraged their strengths to improve treatment experiences and promote retention.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12954-025-01328-3.

Introduction

Opioid use disorder (OUD) is a chronic relapsing condition for which the first-line therapy is treatment with long-acting opioid agonists such as methadone or buprenorphine/naloxone [1, 2]. In Canada, as elsewhere, opioid agonist treatment (OAT) is highly regulated [3], typically delivered within institutional healthcare settings [4], and marked by difficulties in engaging and retaining patients [5–7]. The continually evolving opioid-related overdose crisis [8] has magnified the need to augment the accessibility and attractiveness of OAT [9–13].

Patient-centered care fosters partnerships between patients and providers, empowering individuals and tailoring care to their needs and goals [14–17]. This approach can guide improvements in OAT delivery [2, 18–21]. However, for people who use drugs, substance use stigma shapes healthcare experiences and expectations in ways antithetical to patient-centered care. Stigma undermines people’s sense of legitimacy as patients and breeds mistrust, discouraging engagement in health services [22–28]. These processes are exacerbated when rigid, unfamiliar, or unwelcoming service providers and contexts complicate patients’ ability to appear, self-advocate, and succeed in getting the help they want [24, 25, 27, 29, 30]. Conversely, trusting relationships can humanize medical interactions and help overcome, or modify, systems-based mistrust [24, 25, 31].

Community-based harm reduction organizations take an inclusive and non-judgmental stance towards drug use, working to continuously cultivate trust while addressing clients’ holistic and self-identified needs [25, 32–34]. Harm reduction settings are therefore viewed distinctly from institutional environments, supporting care engagement and patient-centered approaches [23, 25, 28, 35, 36]. However, integrating health services within these settings presents challenges such as securing clinical staff, establishing necessary infrastructure, and reconciling divergent philosophical approaches [32, 37]. Telemedicine, through its remote treatment modality, may facilitate access to virtual healthcare spaces in communities, while simultaneously transforming care logistics, relationships and experiences [38, 39].

This article describes and evaluates a program aiming to offer high quality, patient-centered OAT via telemedicine within a community-based harm reduction site in Montreal, Canada. Using a mixed methods convergent design, we explore how this approach can effectively reach, retain, and serve patients disengaged from OAT and at high risk of drug-related harm. Specific aims were to:

1. Describe patients initiating OAT through the program over two years;

2. Quantify retention rates at one, three, six, and 12 months and describe associated treatment parameters;

3. Quantify rates of testing and treatment for human immunodeficiency virus (HIV) and hepatitis C virus (HCV) within 12 months of OAT initiation; and.

4. Examine patient-reported program experiences to contextualize clinical outcomes and inform on the strengths and limitations of this model of care.

The community-based telemedicine program

Setting and context

Located in Montreal, Quebec, Canada, the program was designed and implemented as a COVID-19 response measure [40]. Procedures were developed shortly after the declaration of a provincial public health emergency on March 14, 2020, and the program began in April 2020. This was a period when pandemic containment measures exacerbated vulnerability to drug-related harm [41–46] but also created opportunities to transform healthcare and harm reduction practices [41, 47–49]. We previously described the context surrounding the program’s development, including regulatory changes supporting implementation [40]. Briefly, these included temporary legal exemptions and new guidelines/recommendations supporting the use of telemedicine for OAT initiation and continuation, as well as “risk mitigation” prescribing of pharmaceutical-grade alternatives to the illicit drug supply [50–56]. Neither program partner had offered either telemedicine for OAT or risk mitigation prescribing prior to the pandemic.

Program partners

The program was co-designed and implemented by the addiction medicine service of the Centre Hospitalier de l’Université de Montréal (CHUM), an academic hospital, and CACTUS Montréal, a community-based harm reduction organization. The CHUM offers an integrated model of care for people with substance use disorders and complex social or medical needs, including a low-barrier OAT program, addiction psychiatry service, and evaluation/treatment for sexually transmissible and blood-borne infections (STBBI). CACTUS Montréal takes a pragmatic and humanistic approach to serving people who use drugs, sex workers, and trans people through prevention, outreach, education, and leisure activities. Services include a needle-syringe program established in 1989 and supervised injection facility operating since 2017.

Both partners are located in downtown Montreal, roughly 450 m apart, and share a longstanding history of collaboration to improve access to health care. This includes initiatives such as deploying CACTUS patient navigators, offering priority appointments at the CHUM, and working with CACTUS outreach nurses to facilitate linkage to addiction and hepatitis C virus (HCV) care, among others. Both organizations are part of larger local networks of community-based services and of people with lived experience, who were informed about the program and able to refer individuals directly to CACTUS for access.

Program description

Guiding principles

The model of care was adapted from the CHUM’s existing low-barrier OAT program and guided by patient-centered and harm reduction philosophies. Fundamental aims were to facilitate access to care, reduce overdose risk, and improve patients’ life conditions. The program targeted individuals otherwise disengaged from OAT and sought to respond quickly and flexibly to their needs. Patient safety and self-defined treatment goals were prioritized in clinical decision making, with no expectation of abstinence. Reducing illicit opioid consumption was however considered an important goal given the increasing local presence of unregulated synthetic opioids (e.g., fentanyl) [42, 43, 57].

Eligibility criteria

Eligible individuals were : [1] using opioids and desired OAT; [2] known to CACTUS; [3] not receiving OAT via another service; and [4] enrolled in the Quebec public prescription drug insurance plan (which covers residents without access to a private plan and those aged over 65 or receiving last-resort financial assistance (i.e., welfare/disability payments)). Exceptions to the third criterion were made in extenuating circumstances, including for individuals expelled or otherwise unable to reliably consult their treatment providers (i.e., unable to attend in-person appointments, reach their provider by phone or internet, or when the provider was on long-term leave).

OUD medications and dosing

Long-acting OAT was a compulsory component of the treatment; patients could choose between methadone, buprenorphine/naloxone, or slow-release oral morphine, in line with contemporary national recommendations [2]. A short-acting oral opioid, typically hydromorphone, was co-prescribed where appropriate and desired by the patient (e.g., to help manage withdrawal, cravings, illicit opioid use). Starting doses, titration schedules, allocation of unsupervised (‘take-home’) doses, and adjustments for missed doses followed national or provincial clinical practice guidelines. Buprenorphine/naloxone was not advised for patients without short-term abstinence goals; short-acting opioids were only offered with this medication in the context of a one-week microdosing induction protocol.

Operating procedures

Specific staff roles and operating procedures are detailed in Supplementary Appendix 1/Supplementary Fig. 1. Briefly, services were delivered from an office space at CACTUS, ensuring confidentiality. A computer with stable high-speed internet was provided, and telemedicine appointments were conducted using CHUM-supplied video conferencing software. The CACTUS program coordinator was responsible for promoting the program, pre-evaluating patients, scheduling appointments, facilitating telemedicine encounters (including establishing the connection, etc.), and providing comprehensive patient support. The medical team, comprising CHUM physicians and nurses, attended appointments remotely from their offices, either together or in separate rooms. The coordinator introduced the team to each patient and, when requested, accompanied patients during appointments. The CHUM physician prescribed and monitored the OUD treatment plan, while the full team addressed patients’ broader health needs. CHUM nurses responded to most patient-initiated requests and coordinated care across services.

Patients were pre-evaluated at CACTUS using a standardized form to determine eligibility and received an intake appointment within 24 h to one week. During this appointment the medical team took a comprehensive patient history and initiated a personalized treatment plan. Patient goals and progress were revisited throughout follow-up and the treatment plan adjusted accordingly. The medical team employed a variety of approaches to increase empathy and trust during interviews, including motivational interviewing. Debriefing with the CACTUS coordinator was also conducted as needed to enhance patient communication. Clinical notes were documented in the CHUM system to ensure the ready availability of medical information and support care continuity and coordination.

To maximise convenience for patients, prescriptions were faxed directly to their nominated community pharmacy. STBBI testing and other procedures requiring direct patient contact (e.g., wound care) were performed by nurses at CACTUS or the CHUM, with patient navigation offered to individuals attending the CHUM. Specimens were sent to the CHUM laboratory for analysis, and results entered into the patient’s medical record. Stabilized patients no longer wishing to attend CACTUS for follow-up could transfer to the regular CHUM outpatient program, transfer to another care provider (e.g., primary care physician, other addiction service), or connect to the medical team via a personal device.

Operating hours

CACTUS was open from 2pm–2am, seven days a week, and did not shut down during the pandemic despite a reduction in service capacity [43]. The telemedicine clinic operated from 1:30 − 3:30pm on Tuesdays and Wednesdays, offering four 30-minute appointments per day. Two clinic days were implemented to ensure all patients could be seen within one week of screening. The early afternoon session was selected to precede syringe distribution hours, thereby supporting privacy and confidentiality, and timed to the coordinator’s availability. Outside official opening hours, patients with urgent or unforeseen health concerns could contact the medical team directly or through the CACTUS coordinator. Appointments with the nurse or physician on duty were scheduled the same day and conducted via the CACTUS platform, by phone, or in person at the CHUM, according to patient convenience. In case of an emergency, patients could attend the CHUM emergency department.

Research methods

Study design

A convergent mixed method design combining retrospective chart review with semi-structured patient interviews was used to evaluate program outcomes and experiences. Quantitative and qualitative data were collected and analyzed independently and in parallel, then integrated at the stage of interpretation. The study was approved by the CHUM research ethics board. The analysis plan was not pre-registered, and results should be considered exploratory.

Quantitative methods

Data were collected from the CHUM electronic medical record (EMR) for all patients initiating OAT between 29 April 2020 (first patient in the program) and 31 March 2022. A standardized abstraction form captured patient characteristics, engagement in OAT and associated treatment parameters, and records of testing/treatment for HIV and HCV. The EMR included pre-evaluation forms completed at CACTUS and data for patients transferred to the regular CHUM outpatient program. Data for patients transferred to another care provider were available until the date of transfer.

Quantitative measures

Patient characteristics were defined using pre-evaluation forms and clinical notes from the intake visit. We assessed patient age; self-identified gender; housing status (stable, unstable [temporary accommodation lacking security of tenure], or homeless [unsheltered or using emergency shelters]); income sources (employment [including undeclared work], government benefits [welfare, disability, pension, emergency response benefits], panhandling, other); registration with a primary care physician (yes/no); duration of opioid use (years); lifetime use of specific opioid classes (heroin, prescription opioids, fentanyl and analogs); past-month opioid injection (yes/no); other past-month substance use (alcohol, powder/crack cocaine, amphetamines, benzodiazepines); past six-month overdose (yes/no); and lifetime receipt of prescribed OAT medications (methadone, buprenorphine/naloxone, slow-release oral morphine, short-acting opioids).

Primary outcome measures were the proportions of patients retained in their initial OAT episode at one, three, six, and 12 months. Retention was defined per Nosyk et al. [58] and other Canadian studies [5, 6, 59, 60] as continuous engagement in OAT with dosing interruptions not exceeding four (methadone, slow-release oral morphine) or five (buprenorphine/naloxone) consecutive days. Procedures for identifying dosing interruptions are described in Supplementary Appendix 2. Treatment parameters, including prescribed medications and doses, were recorded at each juncture.

Defining retention by a single continuous treatment episode fails to capture patients engaged in care despite cyclic treatment interruptions and may obscure some benefits of low-threshold, community-integrated systems with flexible re-entry conditions. We therefore defined as secondary outcomes the proportions of patients with an active OAT prescription at one, three, six, and 12 months post-intake, irrespective of prior treatment discontinuation. Finally, we identified proportions of patients tested and treated for HIV and HCV within 12 months of intake.

Quantitative analysis

We computed descriptive statistics including counts, proportions (categorical data), medians and interquartile ranges (continuous data) in SPSS 29.

Qualitative methods

The qualitative component of the study was conducted between 19 March and 27 May 2021, approximately one year after program initiation, and aimed to explore participants’ experiences and perceptions of the program in the context of the COVID-19 pandemic. A semi-structured interview guide was developed collaboratively with the medical team, CACTUS staff, and a person with lived experience of OUD and OAT. The guide addressed how participants learned about the program, factors influencing their decision to participate, and both positive and negative experiences.

Participants were recruited through convenience sampling. Eligibility criteria required that they were active in the program and able to communicate in French or English. Exclusion criteria included inability to provide informed consent or circumstances in which participation was deemed potentially harmful. All patients were approached and invited to participate unless assessed as ineligible. Participation was entirely voluntary. The CACTUS program coordinator introduced the study to patients during their appointments and scheduled interviews with the research team.

A total of twenty patients completed a semi-structured interview. Due to ongoing COVID-19 restrictions, interviews were conducted at CACTUS over videoconference. Participants provided verbal consent and received $30CAD compensation along with coffee and snacks. Interviews lasted an average of 36 min (range: 15–72).

Qualitative analysis

Interviews were recorded, transcribed, and analyzed using iterative thematic analysis. Transcripts were manually reviewed by members of the research team and an initial codebook was generated. All transcripts were then coded according to the codebook using NVivo. Codes were grouped into themes and an initial thematic tree generated. Themes were then reviewed against the coded data and transcripts and refined through team discussions.

Results

Patient characteristics at intake

A total of 69 people (45 cisgender men, 22 cisgender women, 2 transgender people) initiated OAT through the program between 29 April 2020 and 31 March 2022 (Table 1). Median age at intake was 38 years [IQR: 31–47]. Nearly all patients (96%) had injected opioids in the past month. Over half were homeless (36%) or unstably housed (20%) and less than one-third had a primary care physician (29%). Many patients (44%) reported lifetime use of fentanyl or analogs and 16% had overdosed in the past six months. A further 16% reported an overdose at an unspecified time. One-fifth reported past-month alcohol use (22%) and one-third cocaine or amphetamine use (35%). Income sources were unstated for many patients (41%), but 38% reported income from government benefits and 29% some form of employment. 49 patients (71%) had previously been prescribed OAT; nine had transferred from another service and 21 had received at least two different OAT medications. Overall, 51% of all patients had previously received methadone, 42% buprenorphine/naloxone, and 12% slow-release oral morphine. Only seven patients (10%) had previously been prescribed short-acting opioids with OAT.

Table 1.

Patient characteristics at intake (N = 69)

	Median (Q1–Q3)	n	%
Age (years)	38 (31–47)
Gender
Cisgender man		45	65.2
Cisgender woman		22	31.9
Other gender identity		2	2.90
Housing status
Stably housed		30	43.5
Unstably houseda		14	20.3
Homeless		25	36.2
Income sources
Employmentb		20	29.0
Government sourcesc		26	37.7
Panhandling		5	7.25
Other		3	4.34
None or not stated		28	40.6
Registered with a primary care physician		20	29.0
Duration of opioid use (years)#	9 (1–14)
Lifetime opioid use
Heroin only		8	11.6
Prescription-type opioids only (excl. fentanyls)		17	24.6
Heroin + prescription-type opioids (excl. fentanyls)		13	18.8
Any use of fentanyl or fentanyl analogs		30	43.5
Not reported		1	1.45
Past-month substance use
Opioid injection		66	95.7
Alcohol use		15	21.7
Cocaine/crack cocaine use		15	21.7
Amphetamine use		12	17.4
Benzodiazepine use		6	8.67
Overdose history
Any record of prior overdose		22	31.9
Overdose in past 6 months		11	15.9
Prior opioid agonist treatment (OAT)
Any prior OAT		49	71.0
Methadone		35	50.7
Buprenorphine/naloxone		29	42.0
Slow-release oral morphine		8	11.6
Prescribed short-acting opioid with OAT		7	10.1

^aLiving in accommodation without security of tenure, e.g. motel rooms. ^bIncludes undeclared work (n = 7). ^cIncludes last-resort financial assistance (i.e., welfare or disability payments) (n = 21), Canada emergency response benefits (n = 3), pension (n = 2). ^#26 records missing

Quantitative results: primary outcomes

Retention in the initial OAT episode was 83% at one month, 74% at three months, 67% at six months, and 54% at 12 months. Four patients (5.8%) were transferred to another care provider before 12 months, and one patient died. Table 2 summarizes OAT regimens and doses prescribed at initiation of the first OAT episode (all patients) and at one, three, six, and 12 months (retained patients).

Table 2.

Medication types and median doses (mg/day) at first prescription (all patients) and 1, 3, 6, and 12-month follow-up (patients retained continuously in first care episode)

Long-acting medication	Start: n = 69				1 month: n = 57				3 months: n = 51				6 months: n = 46				12 months: n = 37
	n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)
Methadone	37	(54)	40	(40–40)	30	(53)	65	(50–90)	26	(51)	80	(50–100)	24	(52)	95	(50–110)	20	(54)	98	(50–118)
Slow-release oral morphine	24	(35)	60	(60–60)	24	(42)	300	(200–420)	22	(43)	310	(240–650)	19	(41)	400	(260–820)	15	(41)	450	(260–880)
Buprenorphine/naloxone	8	(12)	5	(0.5–12)	3	(5.3)	20	(13–26)	3	(5.9)	20	(18–26)	2c	(4.3)	20	(20–20)	2	(5.4)	27	(22–32)

Co-prescribed hydromorphonea	Start: n = 69				1 month: n = 57				3 months: n = 51				6 months: n = 46				12 months: n = 37
	n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)		n	(%)	Med (IQR)
With methadone	28	(76)	32	(32–46)	24	(80)	48	(40–64)	22	(85)	52	(36–80)	19	(79)	64	(32–80)	17	(85)	72	(52–84)
With slow-release oral morphine	23	(96)	32	(28–48)	24b	(100)	40	(32–64)	21	(95)	48	(32–56)	19b	(100)	48	(32–56)	15b	(100)	48	(32–64)
With buprenorphine/naloxone	3	(38)	40	(36–40)	1	(33)	32	NA	1	(33)	32	NA	0	(0.0)	NA		1	(50)	48	NA

Med = Median, IQR = Interquartile range, NA = not applicable

^aIndicates the subset of patients also prescribed a short-acting opioid (hydromorphone) and associated doses; percentage denominator is the total number of patients receiving a given long-acting OAT medication

^bOne patient was receiving oral morphine instead of hydromorphone; dosage was multiplied by 5.0 for analysis following recommended conversion factors [61]

^cOne additional patient was receiving subcutaneous extended-release buprenorphine injections (300 mg/month) and hydromorphone (40 mg daily)

54% of patients initiated methadone (median starting dose: 40 mg/day), 12% buprenorphine/naloxone (5 mg/day), and 35% slow-release oral morphine (60 mg/day). Co-prescription of oral hydromorphone was present in 75% of patients initiating with methadone, 38% with buprenorphine/naloxone (specifically, three of four patients following a microinduction schedule), and 100% with slow-release oral morphine. Median hydromorphone doses were 32 mg/day with methadone or slow-release oral morphine, and 40 mg/day with buprenorphine/naloxone.

Proportions of patients receiving each OAT medication and proportions co-prescribed hydromorphone varied little across retention intervals. However, at one month, only three patients retained in their first OAT episode were receiving buprenorphine/naloxone; of eight individuals who initiated treatment with this medication, three were lost entirely to follow-up within one month, one switched to methadone within one month, and one initiated methadone after two months out of treatment.

Doses of both long-acting and short-acting medications tended to increase with duration of retention. From one month to 12 months, median doses of methadone and slow-release oral morphine increased by 50%. Median hydromorphone doses were higher for those prescribed methadone relative to slow-release oral morphine at one month, and increases were more pronounced by 12 months (50% vs. 20%).

Quantitative results: secondary outcomes

Figure 1 presents proportions of patients with an active OAT prescription at one, three, six, and 12 months post-intake, including treatment re-initiation. Overall, 87% of patients had an active prescription at one month, 83% at three months, 77% at six months, and 71% at 12 months. Excluding patients transferred to another provider, these proportions were 87%, 84%, 79%, and 75%. At each follow-up, roughly 55% of active prescriptions were for methadone, 40% slow-release oral morphine, and 5% buprenorphine/naloxone. Co-prescription of hydromorphone ranged from 83% at one month to 88% at 12 months.

Fig. 1.

Proportion of patients retained, and long-acting medication type at 1, 3, 6, and 12 months post-intake. a Bars: Proportion of patients retained continuously in the first care episode, with an active prescription despite prior discontinuation, and transferred to another care provider at 1, 3, 6, and 12 months post-intake; and b Lines: Long-acting medication type among active OAT prescriptions at 1, 3, 6, and 12 months post-intake

Of 69 patients, 48 (70%) were tested for HIV within 12 months of intake. A further five (7.2%) had a recent HIV test result (< 6 months old) in the EMR at intake. Of three known HIV-positive patients, one had an undetectable viral load at intake; one initiated HIV treatment within 12 months; and one did not initiate treatment despite multiple attempts at linkage to care.

A total of 50 patients (72%) were tested for HCV antibodies or RNA within 12 months of intake. A further five (7.2%) had a prior antibody-positive result in the EMR, including two with a recent RNA-negative result. Supplementary Fig. 2 presents a complete HCV cascade of care with associated definitions. Briefly, of 33 patients identified as HCV antibody-positive, 27 (82%) were tested for HCV RNA and 15 (45%) were found RNA-positive. Of these, 13 (87%) were prescribed HCV treatment and 10 (67%) had evidence of sustained virologic response in the EMR.

Qualitative results

Semi-structured interviews were completed with eleven cisgender men, eight cisgender women, and one transgender woman (N = 20; Supplementary Table 1). Seventeen had prior experience of OAT and eight were unstably housed or homeless at intake. Interviews took place a median of 163 days (IQR: 44–302) after intake. Ultimately, sixteen were continually retained in OAT for 12 months. One was lost-to-follow-up after six months and three ceased treatment for periods exceeding 30 days.

Six main themes emerged from the interviews, reflecting participants’ experiences of: (1) the community-based treatment setting, (2) the multidirectional collaboration between program actors, (3) the expanded range of medication options, (4) the use of telemedicine, (5) the implication of community pharmacies, and (6) demarginalization, engagement, and renewal. Quotations have been translated from French and truncated for brevity, where necessary, and medication types substituted for trade names.

Community-based setting

Implementation within a community-based harm reduction setting emerged as a key factor supporting patient engagement and retention. Participants highlighted benefits such as raising awareness of the service, enhancing convenience, destigmatizing care, and leveraging existing relationships of trust. Some emphasized the practicality of accessing care in a setting they frequented for other needs:

CACTUS is a place that’s important to me. You know, I go there every day to inject. And the fact that the service is offered there, I find it perfect. It’s just – it fits really well into my routine (6: Woman, aged 25–34)​

The warm, non-judgmental, caring, and respectful welcome at CACTUS was also crucial. This approach and the positive relationships it fosters strongly influenced clients’ experiences of seeking and accepting care:

I had a relationship of trust, because I’ve known them for years… If [the CACTUS program coordinator] proposed something to me, they’ve known me for years, I knew that it could just be beneficial for me. Otherwise, they wouldn’t have offered it (16: Man, aged 45–54)

A ‘collaborative triangle’ between program actors

Multidirectional collaboration among patients, CACTUS workers, and the medical team was evident across various aspects of the treatment experience. First, alliance between the health and community sectors enabled a swift response connecting people at risk of harm with care:

I was having problems. I was using a lot, then [the coordinator], it was her who spoke to me about it. Then, you know, it’s easier, she accompanied me (3: Man, aged 35–44)

CACTUS staff further enhanced the treatment experience by offering an accessible and holistic support system tailored to patients’ needs. Participants described receiving emotional, instrumental, informational, and appraisal support from the program coordinator and emphasized the value of feeling accompanied during their treatment journey:

What I appreciate is the speed and promptness with which my questions are always answered. And the fact that, if it’s been a while since we’ve had any communication, [the coordinator] will follow up with me, say hello… There’s this human aspect which is magnificent in this program (2: Woman, aged 55–64)

Collaboration was also evident in relationships between patients and the medical team, especially regarding medication and dosing decisions. Participants conveyed a strong sense that their experiences were acknowledged and considered:

If we tell [the doctor] something, he’s not going to act like it’s not important or that we’re bullshitting or whatever. He’ll really listen to us. And then, he’ll adapt the treatment according to what we tell him (6: Woman, aged 25–34)​

Many emphasized an atmosphere of warmth, empathy, respect, and non-judgment in their medical interactions, alongside a perception that the medical team was competent, knowledgeable about the realities of people who use drugs, pragmatic, and genuine in their approach:

I find that the doctor, he’s really understanding, he knows what he’s doing. And it’s just the person, who you know – you don’t feel like a druggie in front of him. He really understands what it is, what you’re going through (4: Woman, aged 35–44)

Flexible and non-punitive procedures were contrasted with prior experiences of OAT, further contributing to a sense that patients and the medical team could collaborate effectively:

The other doctors I’ve had before, they were stricter, less flexible. You know, less – how can I say it – “open-minded”. They’re tougher… I always felt like I needed to lie to the doctor or something. Whereas now, I just tell him the truth and it doesn’t cause problems (7: Man, aged 35–44)

If ever you miss your appointment, well, [the doctor] prescribes for you anyway then sees you the following week… Because other places if you don’t go to your appointment, bam! You’re cut off in one shot. Then, you fall back into withdrawal (16: Man, aged 45–54)

Expanded medication options

An expanded range of medications was a drawcard for many participants. Co-prescription of hydromorphone motivated some to engage and remain in the program, addressing priorities such as managing physical pain, withdrawal, and the cost or risks associated with illicit opioid use:

Well, the [hydromorphone] is definitely… an incentive or a motivation to stay in treatment and to go there every day because you get the morphine [sic] that you can take as you wish… At least you know what you’re taking, you know the quantity, you know the effect it’s going to have. Whereas the drugs you buy in the street, you never know what you’re going to get (9: Man, aged 35–44)​

Interviewees did not seem to object to receiving a compulsory OAT ‘backbone’. One participant, seeking only hydromorphone, mentioned that this requirement delayed his entry to the program. Others explicitly noted benefits of their long-acting medication. Hydromorphone co-prescribing nevertheless enhanced the perceived flexibility and desirability of the treatment plan:

[The buprenorphine-naloxone I received in the past] isn’t something that interested me much. I just didn’t want to be in withdrawal… With my [hydromorphone] and [slow-release oral morphine], I want to go to the pharmacy to get it (10: Man, aged 25–34)

While methadone remained popular among the medications on offer, interviewees rarely discussed this choice; perhaps due to a lack of novelty in the treatment landscape. Conversely, some participants emphasized the appeal of trying slow-release oral morphine following prior adverse experiences with methadone or buprenorphine/naloxone:

I was motivated to try a new treatment because the others I’d tried in the past, several times each they hadn’t worked… Methadone made me sick, literally more than if I didn’t take it. [Buprenorphine/naloxone] gave me terrible liver pain… So, let’s just say [treatment with slow-release oral morphine and hydromorphone] is practically magic, in fact (20: Woman, aged 25–34)

Use of telemedicine

Participants expressed mixed feelings about receiving care via telemedicine. On one hand, the modality was viewed as uncomplicated and practical, and appreciated for its efficiency:

I find it goes super well compared with trying to see a doctor in person. That’s long, it’s complicated. Whereas this was easy. I think it’s the best way there is (11: Man, aged 45–54)

Some participants nevertheless expressed regret, unease, or frustration at the lack of opportunity for in-person conversation and physical examination. This could make the experience feel cold (like “you’re still a number, a file on a computer”) and raised questions about the limits of telemedicine:

Me, I say to myself, I’ve got a sore arm, but you can’t check my arm… For little boo-boos, or if you say you’re in withdrawal and everything, I find it perfect. But I think there’s a limit at some point (3: Man, aged 35–44)

Finally, engaging with telemedicine requires specific equipment and skills and may demand a degree of adaptation. For some patients, it is essential to provide not only the technology but also support in using it, such as help connecting to appointments and managing the microphone and camera:

I’m not very tech-savvy… That adaptation is a bit difficult, but then, it’s [the coordinator] who takes care of that. But the technological aspect scared me a little at first (2: Woman, aged 55–64)

Implication of community pharmacies

The use of telemedicine did not address some barriers to OAT participation, including a requirement for frequent pharmacy attendance. Pharmacy visits could be inconvenient and unpredictable, interfering with participants’ efforts to make positive life changes:

What’s negative is that I’m always at the pharmacy. I’ve been – you know, at one point I missed Christmases, I missed doing things with my family because I was stuck at the pharmacy every day. It’s kind of hard (3: Man, aged 35–44)

While some participants described pharmacy staff as friendly, helpful, and engaged partners in their care, others reported interpersonal conflicts and practices that might be perceived as exclusionary or stigmatizing:

I find it a bit irritating, the door out the back [for OAT recipients]… I find it irritating because I feel like – then I pass hidden by the back door, you know, to hide. But that’s also practical. I don’t wait long, and the pharmacist is there just for that (17: Man, aged 35–44)

De-marginalization, engagement and renewal

Overall, participants felt surrounded by an extended and cohesive team dedicated to addressing their various needs and helping them achieve personal goals at their pace. This sense of inclusion and genuine concern for the patient’s wellbeing and autonomy was transformative for a group that often experiences marginalization and ostracism:

No matter how, there’s somebody who’s there for you. You can count on whoever for whatever in the program: when everything’s going well, when you relapse, when you have dark thoughts… It’s important when you want to get through it that there are people around you. It’s boring when you’re all alone, so you ruin your life (12: Woman, aged 25–34)

Positive experiences in the program offered an opportunity, however tenuous, to restore their eroded trust in health services. Rather than generalizing this trust broadly, participants tended to concede that exceptions could exist in an otherwise flawed system:

(Interviewer: What was your perception of health services, before entering the program? ) Well, a system that’s broken, that’s not humane then very, very stigmatizing and I don’t know, I don’t know if I can describe it better than that. ​(Interviewer: Okay. Since you’ve been in the program, has your opinion changed a little?​) Yes, absolutely. I think it’s, depending on which door you enter the hospital through, in which state, which professional you come across, it’s going to have a big influence on the experience there, clearly (20: Woman, aged 25–34)​​

For some, the program paved concrete pathways to health service engagement, with participants benefiting both from the extensive infrastructure of the CHUM and the support of program actors in navigating health system transitions:

I’m well surrounded. It’s good. It’s really, really good. I had an appointment on the 11th floor of the CHUM for a general examination… They offered me blood tests, the test to see if you’re diabetic, your general health. It’s been 10 years since I put my name down to get a family doctor. I’m still waiting. But now, I have a doctor (15: Man, aged 55–64)​

Participants reported significant improvements in their quality of life and self-concept as they stabilized or reduced their drug consumption. Several equated managing opioid addiction through the illicit market to a full-time job; a vicious cycle in which they were constantly chasing money or drugs to stave off withdrawal. The ability to devote their time, money, and energy to other pursuits offered a sense of renewal and newfound possibility:

The program absolutely changed my life. You know, I’m more functional. So, my self-esteem has increased. I have more confidence in myself. I can – I’m looking for an apartment, a job… I got in contact with my whole family, with my kids (11: Man, aged 45–54)

Discussion

This mixed methods study combined a retrospective chart review with semi-structured interviews to investigate patient outcomes and experiences of a low-barrier OAT program implemented within a community-based harm reduction setting via telemedicine. Overall, findings suggest the program succeeded in engaging structurally vulnerable individuals at high risk of opioid-related harm but disengaged from OAT, and delivering care within a patient-centered framework.

Patients enrolled in the program were typically injecting opioids and many reported potential barriers to successful treatment including unstable housing and use of fentanyl, stimulants, or alcohol. All were disengaged from active care and one-third reported a prior overdose at the time of enrolment. Meanwhile, 70% had previously received OAT, nearly half of whom had tried at least two different long-acting agonists. It is crucial to investigate how to improve outcomes for patients who have engaged in multiple pharmacological treatments without success and are at high risk of treatment complications and abandonment [62–64]. Our results unveil various programmatic features that may have facilitated retention and fostered a variety of positive outcomes for such individuals.

Overall, qualitative data suggest the program’s success was underpinned by trust, respect, and the alignment of practices with patient-centered care and harm reduction principles. These include individualizing care to patients’ unique needs, goals, and preferences; developing trusting and collaborative therapeutic relationships; and addressing needs within a holistic (e.g., biopsychosocial) framework [14, 16–18, 65]. Structural and system features that can support patient-centered care include welcoming and accommodating therapeutic environments; skillful, cohesive and effective teams; and flexible systems that enhance service navigability and coordination [15–17]. For people who use drugs, implementing non-punitive and non-stigmatizing practices that acknowledge and respond to the reality of ongoing drug use may be particularly important [11, 30, 35, 66–69].

In the present study, integrating treatment within a community-based harm reduction setting facilitated patient-centered care by enhancing accessibility and convenience, providing a safe and familiar care environment, and improving the coordination of medical and non-medical care and support [11, 15, 17, 18]. Trust in community workers and a history of supportive interactions in this setting were critical factors promoting patient engagement [24, 25, 36]. The medical team’s attunement to harm reduction principles was then essential in further blurring institutional boundaries and cultivating positive therapeutic relationships. Participants emphasized the team’s interpersonal qualities (e.g., empathy, non-judgment, authenticity, active listening), clinical proficiency, and pragmatism within a flexible and non-punitive treatment framework as key factors supporting their ongoing engagement, often contrasting these against prior OAT experiences [11, 15–18, 70].

Employing strict criteria for OAT discontinuation [58], 54% of patients were retained in their first treatment episode for a full year. Retention thus compared very favorably to large-scale Canadian studies using the same criteria; across 70 addiction treatment centers in Ontario, 12-month retention was 32% for methadone and 20% for buprenorphine/naloxone between 2014 and 2021 [71]. In British Columbia, province-wide 12-month retention was 18% for methadone and 7.9% for buprenorphine/naloxone between 2008 and 2018, and 8.9% for slow-release oral morphine between 2017 and 2018 [59]. Taking a pragmatic view, 71% of our patients had an active OAT prescription through the program after 12 months, indicating their retention in a system of care despite interruptions. This represents an important outcome from the perspective of increasing overall OAT coverage and ensuring that systems support patients requiring multiple treatment attempts [72–76].

A diversity of medication options and collaborative approach to treatment planning were foundational to achieving patient satisfaction and individualized care [11, 18, 75, 77, 78]. Short-acting opioid prescriptions, included in the majority of treatment plans, responded directly to patient priorities such as managing pain or withdrawal symptoms and reducing reliance on illicit drug markets. This offering positioned treatments as flexible and desirable, encouraged ongoing participation in the program [79, 80], and shaped perceptions of the medical team as pragmatic and open-minded. Meanwhile, the selection and dosing of long-acting opioid agonists emerged as critical junctures for shared decision making and relationship-building [69, 81–83]. While methadone was the most commonly prescribed, slow-release oral morphine was also popular and highly valued by some patients [63, 84, 85]. Of note, only eight patients chose to initiate treatment with buprenorphine/naloxone, which was offered with adjunct short-acting opioids during induction only, and just three were retained on this medication for one month. This suggests a need to better understand why individuals rebuff this regimen, ensure close monitoring, and offer alternative treatment options before disengagement occurs, including co-prescription of short-acting opioids alongside buprenorphine-naloxone [2, 86–88].

A focus on trust lends deeper insight into our findings. Trust represents an expectation of positive outcomes under conditions of vulnerability or uncertainty and, in medical settings, requires confidence in a provider’s motivations or intentions as well as their skill or competence [24, 70, 89]. These beliefs develop through socially contextualized interactions, during which patients gather and interpret cues to define their relationships with providers [68, 69]. Structural factors and past experiences shape interactions by influencing expectations and interpretations, as well as the bounds of possible negotiation [68, 89, 90]. Positive or negative feedback loops may occur as expectations and perceived experience influence each other cyclically [70]. For these reasons, trust is eroded by the manifestation of stigma in unwelcoming and de-legitimating healthcare spaces, interactions, policies, and practices [25, 27, 29].

Our study suggests, first, that trust accumulated in community-based organizations or workers can extend to affiliated treatment programs, helping overcome a general distrust of medical providers or institutions [24, 70]. Localizing programs within community settings then reconfigures the socio-spatial context of treatment delivery, promoting feelings of comfort and legitimacy [24, 25, 29, 91]. Together, these factors may reshape expectations and perceptions of the treatment experience to promote trust-building [70, 89], provided that approaches align with community values [32, 37]. Second, clinical practices assume symbolic meaning and appear to influence trust by configuring patient-provider subject positions [69, 91–93]. Participatory decision-making, flexible treatment structures, and proactive stances toward patient needs signal respect and concern rather than power and control, facilitating rapport-building and honesty [24, 25, 31, 68, 69]. Finally, positive treatment experiences may revise or diversify meanings attributed to the medical profession, helping restore trust in healthcare institutions [70, 94]. In our program, roughly 80% of all patients were assessed for HIV and HCV infection, and nearly 90% of those testing positive for HCV RNA initiated treatment; an illustration of engagement in care and a promising finding given the challenge of HCV elimination [95–97]. Additionally, many patients eventually transferred their follow-up to the regular hospital outpatient clinic, indicating some willingness to re-engage with mainstream services.

Access to OAT expanded in Montreal as COVID-19 restrictions eased, with an increasing number of programs incorporating short-acting opioid as “safer supply” medication. Despite this broader availability of harm reduction–based OAT, our program remains active and continues to engage individuals who are disaffiliated from traditional OAT providers. This suggests that, even within a comprehensive system of care, community-based organizations play a crucial role in reaching new patients and re-engaging those who have become disconnected [98].

Participants viewed telemedicine as both beneficial and constraining; a practical and convenient means of accessing care that could also feel impersonal or limited in its application. The bearing of this technology should however be considered in relation to the broader patient experience, including its contribution to reshaping the context, arrangements, and relations of care as just described [39]. Thus, while a social and physical distance of remote encounters was noted, participants nevertheless emphasized the ‘human aspect’ of the program and its responsiveness to their various medical and emotional needs.

Our program did not alter the need to regularly collect medication from pharmacies. Community pharmacies play an important role in OAT delivery in Canada and may require greater attention in efforts to improve the treatment experience [99–101]. Meanwhile, dispensation and monitoring technologies (e.g., biometric machines) may improve treatment flexibility and overall outcomes [102, 103].

Study limitations should be noted. First, quantitative data were collected exclusively via the CHUM EMR and may be incomplete, particularly for variables not relating directly to OAT provision. Ethnicity and race were not documented, and it is possible that results are not generalizable to particular subgroups such as racialized and Indigenous populations. We were also unable to determine loss-to-follow-up between the stages of pre-evaluation and the intake visit, as this information was not systematically collected. Second, qualitative participants were recruited via the CACTUS program coordinator and many were retained continuously in treatment for one year. Findings may therefore overrepresent positive treatment experiences and relationships. Our research design did not incorporate gender-based analyses and may obscure underlying differences in program effectiveness or experiences.

Finally, our program was initiated for a specific population during the COVID-19 pandemic, operating under risk mitigation guidance supporting emerging alternative OAT and safer supply models. Although short acting opioid medications were offered and some participants emphasized these as an important aspect of the program, we did not evaluate their specific contribution to patient outcomes. Safer supply programs implemented in Canada vary widely in terms of populations served, eligibility criteria, and treatment approaches. There is an urgent need for a comprehensive research agenda to characterize these models of care, evaluate their efficacy and impact, and generate evidence-based guidance and recommendations accordingly. Our study contributes to this emerging evidence base and underscores the need for further research into the role of safer supply within OAT delivery.

Conclusion

Our community-based telemedicine program presents an alternative model of OAT delivery that has achieved excellent retention and transformative treatment experiences for patients. When underpinned by a strong spirit of collaboration between community and medical experts, telemedicine represents an efficient means to bridge these sectors and overcomes many logistical challenges associated with delivering OAT outside traditional clinical settings [37]. Our program demonstrates that providing people with technology and support to access telemedicine within community settings complements existing services for those disaffiliated from the health system. Such low-threshold, patient-centered initiatives are an important component of a care continuum capable of engaging and serving individuals at high risk of opioid-related harm. Programs should employ harm reduction approaches and aim to provide holistic care coordination and support, while ensuring capacity to implement individualized treatment plans within a non-judgmental, respectful, and collaborative therapeutic environment.

Supplementary Information

Below is the link to the electronic supplementary material

Abbreviations

CHUM

Centre Hospitalier de l’université de Montréal

EMR

Electronic medical records

HCV

Hepatitis C virus

HIV

Human immunodeficiency virus

IQR

Interquartile range

Mg

Milligrams

OAT

Opioid agonist treatment

OUD

Opioid use disorder

RNA

Ribonucleic acid

STBBI

Sexually transmitted and blood-borne infections

Author contributions

JB, SBH and RCF: conceptualization, design, methodology, investigation, writing, review and editing; SBH: original first draft writing; AS: Conceptualization, review and editing, project administration; SC, GBL: investigation and resources; SKL, RL: investigation; CDM: investigation, writing, review and editing. JB: supervision and funding. all authors reviewed the manuscript.

Funding

This work was supported by the Canadian Institutes of Health Research [grant numbers REN-181678; CUG-171602; 950-232703].

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study was approved by the Centre Hospitalier de l’Université de Montréal (CHUM) research ethics board. All participants signed an informed consent to participate in the study. The analysis plan was not pre-registered and results should be considered exploratory.

Consent for publication

Not applicable.

Competing interests

JB reports a relationship with Gilead Sciences that includes: consulting or advisory and funding grants and a relationship with AbbVie that includes: consulting or advisory, outside the scope of this manuscript. All other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.