1 INTRODUCTION 1.1 Background

Chronic pain is a major public health issue affecting one in five adults in North America1-3 and leading to significant negative impacts on patients' daily living.3, 4 Pharmacological therapies are prescribed to a majority of patients (62%–84%)5-7 as part of the recommended multidisciplinary biopsychosocial approach to chronic pain treatment.8, 9 In North America, the most commonly prescribed medications to manage chronic pain are opioids, nonsteroidal anti-inflammatory drugs (NSAIDs), acetaminophen, antidepressants, anticonvulsants and muscle relaxants.5, 10-12

Chronic pain pharmacotherapies are highly beneficial for patients in terms of pain relief and quality of life. However, most analgesics can lead to adverse effects, which may intensify when several prescribed and/or over-the-counter medications are combined.13, 14 For example, commonly prescribed medications to treat chronic pain, such as opioids15 and antidepressants,16 frequently cause nausea, dry mouth, drowsiness, constipation, dizziness and/or headache.15, 16 Furthermore, clinically serious adverse effects of long-term opioid therapies, such as opioid use disorder and overdose-related respiratory arrest, have been widely described in the past few years.17-19 NSAIDs, for their part, increase the risks for serious cardiovascular and renal diseases.20-22 In addition to being detrimental to patients' physical health condition, adverse effects often lead to negative consequences on quality of life11, 23 and can deteriorate the psychological condition.24 Preventing and managing adverse effects of analgesics and associated risks can thus lead to significant challenges for both patients and clinicians, especially for pharmacists, who play an essential role in treatment optimization as part of a primary care approach to chronic pain.25-27 In several Canadian jurisdictions, pharmacists' scope of practice has been recently extended. They can now adapt or manage prescriptions and independently prescribe medications under certain conditions.26, 28 Therefore, they are more than ever at the front line to monitor, prevent and manage adverse effects of chronic pain pharmacotherapies.

Available data on analgesic adverse effects come primarily from randomized clinical trials, which makes these data difficult to transfer to real-life experiences.29 Though the nature and frequency of these adverse effects have been examined in experimental settings, little is known on the experiences and concerns of people living with chronic pain and pharmacists regarding adverse effects and their management. Indeed, the most frequent or clinically serious adverse effects may not be the most problematic in the daily experience of patients and pharmacists. The few studies examining patient and clinician perspectives regarding chronic pain pharmacotherapies focused on opioids only and did not include pharmacists.30-33 One qualitative study underscored that patients and physicians had distinct preferences regarding the management of opioid-induced constipation.30 Furthermore, several studies highlighted shared concerns between patients and physicians regarding risks of serious adverse effects, such as opioid use disorder and overdose, especially in the context of the ongoing overdose epidemic.32, 34-39 However, studies showed that divergent perspectives on opioid tapering were challenging for patient–provider communication and could result in increased patient stigmatization as well as negative impacts on pain outcomes.38, 40-42 Currently, there is no research examining the experience and management of the adverse effects of diverse chronic pain pharmacotherapies from the double perspective of patients and pharmacists. Such research is highly needed to understand patients' and pharmacists' support needs in the daily management of pain and drug adverse effects.

1.2 Objectives and theoretical framework

The objective of this study was to understand the experiences and decision-making of people living with chronic pain and community pharmacists regarding the adverse effects of analgesics. More precisely, the study sought to understand: (a) how, and according to which criteria, people with chronic pain and pharmacists define and classify the adverse effects of pain medications; and (b) how people with chronic pain and pharmacists make decisions to prevent and manage the adverse effects of pain medications in their daily life or practice.

The concept of adverse effects used in this study included any unfavourable, unpleasant or harmful outcome following the intake of an analgesic drug or the interaction between an analgesic drug and other medications.43, 44 Adverse effects are distinct from the broader concept of side effects referring to all outcomes that do not pertain to the drug's primary therapeutical goal, be they considered adverse, neutral or beneficial.43

This study was based on a qualitative approach to the social experience of chronic illness and the provider/patient relationship, as theorized by interactionist health sociology.45-49 More precisely, this study was inspired by the theoretical framework of the ‘meaning of medication' defined by Conrad.50 This social-constructionist approach proposes an alternative analysis of adherence by focusing on patients' daily experience of chronic medication intake. It offers a person-centred perspective on the management of chronic illness and patients' adaptive strategies to integrate the treatment into their daily living.46, 50 According to Conrad, it is essential to analyse patients' diverse medication usages independently from their conformity to medical prescription, to understand patients' strategies for living with—or in spite of—their adverse effects.50 Such an approach enabled us to examine the criteria people with chronic pain utilized to classify the effects of their medication as adverse effects, and the strategies they deployed to manage these adverse effects, including treatment modifications. This theoretical framework also supported our understanding of pharmacists' experiences through the analysis of how the management of analgesic adverse effects was included in their daily practice and influenced by interactions with their patients through their convergent and divergent views. Such a symmetrical approach to patients' and pharmacists' experiences intends to open avenues for improving communication in the clinical relationship to optimize the management of adverse effects.

2 METHODS

The methods are reported following the Standards for Reporting Qualitative Research (SRQR) recommendations.51

2.1 Design and recruitment

This was a qualitative study using online (Zoom™) focus groups (FGs) with people living with chronic pain and community pharmacists. Data were collected in the province of Quebec, Canada, between July 2020 and February 2021. The research team conducted six FGs with people living with chronic pain and six FGs with pharmacists. Each FG comprised three to five participants, as recommended in the literature for online FGs.52, 53 The study was approved by the Research Ethics Boards of the Centre hospitalier de l'Université de Montréal and the Université du Québec en Abitibi-Témiscamingue. All participants provided written informed consent by email and all data were anonymized.

Participants living with chronic pain were recruited in a provincial web-based prospective cohort study examining pharmacological and nonpharmacological treatments used by adults living with chronic pain.54 Additional information about the cohort study and recruitment procedures is shown in Figure 1. For this qualitative study, eligibility criteria included: (a) living with pain for at least 6 months, (b) using pharmacological pain treatments, and (c) experiencing one or more moderate to severe adverse effects. Among those consenting to be contacted for other studies, 150 eligible individuals were randomly selected with sex-based stratification. They were sent an email offering them to participate in an FG. The study coordinator scheduled appointments with those who answered positively until all FGs were complete. A total of 26 people with chronic pain were recruited (Figure 1).

Recruitment procedure for participants living with chronic pain. *The Chronic Pain Treatment (COPE) cohort study included individuals meeting the following eligibility criteria: (1) reporting persistent or recurrent pain for more than 3 months; (2) being at least 18 years of age; (3) being able to complete a web-based questionnaire in French and (4) living in the province of Quebec. **The email offering to participate in the focus group was sent to 50 cohort participants in June 2020, 50 cohort participants in July 2020, and 50 cohort participants in August 2020. ***Two participants expressed their interest to participate but were not enrolled because they were unavailable on the scheduled date for the last focus group (FG)

As FG guidelines recommend internal homogeneity of participants,55 the study coordinator scheduled distinct groups for participants who used opioid treatments and those using nonopioid treatments. Indeed, as several studies highlighted the strong stigma surrounding opioids,39, 56, 57 it was essential to ensure that the participants would be comfortable talking about their treatment without worrying about being judged by their peers. Participants were asked which types of treatments they were using during the recruitment, to include them in the appropriate group. Three FGs were conducted with people using opioid treatments and three with people using nonopioid medications.

To be eligible, pharmacists had to work in community pharmacies and/or family medicine groups in the province of Quebec. Family medicine groups are multidisciplinary primary care facilities offering services by general practitioners, pharmacists, nurses and/or social workers.58

Pharmacists were recruited using diverse strategies, including announcements in professional associations' newsletters, websites and/or social media pages, invitations shared in the researchers' professional network and the snowball method. Among the 20 pharmacists who accepted to participate, 19 were finally included (one cancelled participation due to insufficient time). Sociodemographic characteristics of participants are described in Tables 1 and 2.

Table 1. Participants' characteristics: Individuals living with chronic pain Characteristic N Total number of participants with chronic pain 26 Age 30–39 1 40–49 6 50–59 6 60–69 10 ≥70 3 Gender Female 12 Male 14 Ethnicity White 24 Indigenous/First Nations 2 Average pain intensity in the last 2 weeks Mild (1–3/10) 2 Moderate (4–6/10) 17 Severe (7–10/10) 7 Pain duration (years) 1–10 2 11–20 8 21–30 8 31–40 4 >40 4 Origin(s) of paina Accident 7 Disease 4 Trauma/repeated movement 5 Pregnancy/childbirth 2 Undetermined 9 Pain location(s)a Multisite/generalized 14 Head/face 13 Neck 17 Shoulder 16 Arm/elbow/wrist 14 Hand 11 Back 24 Chest/rib 5 Abdomen 6 Hip 13 Buttock/genitals 8 Leg/ankle 15 Knee 15 Foot 15 Provider(s) prescribing pain medicationa Family physician 16 Multidisciplinary pain treatment clinic 7 Other specialist 5 Work status Employed 6 Unemployed 4 Invalidity pension/paid sick leave 7 Sick leave without financial compensation 2 Retired 7 Annual income <CDN $20,000 11 CDN $20,000 to CDN $40,000 5 CDN $40,000 to CDN $60,000 5 CDN $60,000 to CDN $80,000 3 >CDN $80,000 2 Table 2. Participants' characteristics: Pharmacists Characteristic N Total number of pharmacists 19 Age 25–29 7 30–34 7 35–39 4 55–59 1 Gender Female 11 Male 8 Ethnicity White 16 Asian 2 Black 1 Type of practice Community pharmacy only 8 Community pharmacy and family medicine group 11 Years of practice 0–4 6 5–9 8 ≥10 5 2.2 Data collection

FGs lasted between 90 and 120 min, with a 10-min break for participants' comfort. FGs were conducted by two experienced research assistants with backgrounds in psychology and sociology. One of them moderated the discussion while the other one observed, took notes and assisted participants with technical issues when needed. Each participant was given a pseudonym to protect anonymity during discussions. FGs were audio-recorded and transcribed verbatim. Data collection was supervised by the study's principal investigators. After the FGs, participants completed an online sociodemographic questionnaire, and they received CDN $75 for their time.

The semistructured discussion guides included open-ended questions pertaining to participants' definition and management of adverse effects (see Supporting Information Material: discussion guides for patients and pharmacists). The moderators used prompts to help participants develop their narratives and make sure all of them had a chance to talk. The moderators did not establish a preliminary list of adverse effects, allowing participants to decide which effects they wanted to include in this category.

2.3 Data analysis

Data analysis was driven by Corbin and Strauss' grounded theory methodology.59 The analysts went back and forth between data collection and analysis to be the closest possible to theoretical saturation. The research team used several strategies to assess and foster saturation throughout the study. First, after each FG, both moderators wrote detailed reports documenting both surprising/new and redundant elements discussed by participants. Furthermore, the analysts started analysing the data concurrently with data collection. Team meetings after each FG enabled them to adapt prompts in the forthcoming FGs to verify analytic ideas and to explore unanticipated topics emerging from participants' discussions (e.g., adverse effects of natural products and cannabis). In addition, the initial research protocol included a possibility for conducting additional one-on-one interviews with new participants in the case 12 FGs would not provide sufficient data. We finally did not use this strategy as we could develop a consistent analysis using the data from the FGs.

In this study, the analysts adopted an interpretivist epistemological positioning considering that analytic categories or themes are coherent constructs resulting from the researchers' interpretations of participants' narratives.59, 60 Three members of the research team were involved in data analysis and code development. The involvement of three analysts coming from various disciplinary and professional backgrounds (social sciences, community health, clinical psychology) offered diversified insights into the data, fostering reflexivity and enriching interpretations.

The analysts used an iterative and reflexive approach combining attention to the research objectives with their inductive interpretations of data. We used Corbin and Strauss' constant comparative method with open and axial coding strategies to develop both a comprehensive understanding of each FG and a comparative perspective on the data.59 NVivo-12 Software61 supported data storage and management. Memo writing throughout the analysis process helped us to progressively build the analytic categories. Regular team discussions supported the constant comparison procedures and allowed us to produce an integrated interpretation of the data set. Given that participants were francophones, a professional translator translated all cited quotes and the discussion guides for this article. The translation was double-checked by the research team.

3 RESULTS

This section centres on (1) patients' and pharmacists' experiences and dilemmas while defining, classifying and prioritizing analgesic adverse effects; (2) patients' and pharmacists' practices, expectations and challenges in the management of analgesic adverse effects. A graphical representation of the analytical categories and their dimensions is provided in Figure 2.

Swallowing the pill of analgesic adverse effects

3.1 Defining and prioritizing adverse effects: An individualized benefit-risk balance 3.1.1 Adverse, neutral or beneficial: Individualized classifications of side effects People with chronic pain and pharmacists defined and prioritized adverse effects according to their perspective on the benefit/risk balance of pain medication intake. Participants actively constructed the notion of ‘adverse effect’ through the classification of the different effects as either undesirable/adverse, tolerable or desirable. For both patients and pharmacists, this classification resulted from their individual assessment of medication benefits and risks, which varied according to diverse characteristics, such as the patient's age, comorbidities, professional activity and familial situation. The importance of personalizing the treatment was recurrent in FG discussions. Depending on the patient's situation, a given effect could be alternatively defined as an adverse, undesirable effect or as a desirable, helpful effect. For example, some pharmacists explained that drowsiness was a positive side effect for patients suffering from insomnia, a minor adverse effect for retired patients, while it was concerning for patients with a professional activity requiring productivity or driving:

Sometimes, I recommend a pain medication and I say, ‘It's going to make you drowsy’. Some patients say, ‘No problem. I'm retired. I'm home all day’. But some others who have to work, who are on the road, truckers, for them it's just impossible. So, their occupation and schedule can make a difference when choosing a treatment. (Pharmacist, Part.38-FG10)

According to several patients, some side effects of their pain medication positively impacted their other health conditions:

I have fibromyalgia, severe generalized arthritis and emphysema, and the medication allows me to be active. I can go for a walk, I can concentrate enough to read a book, which I wasn't able to do before because of my ADHD [Attention Deficit/Hyperactivity Disorder]. (Patient, Part.7-FG2)

3.1.2 Present quality of life versus risks for future health

Participants appraised medications' benefit/risk balance through two temporalities: present quality of life and risks for future health. Participants hierarchized adverse effects according to the importance they gave to each of these temporalities in the process of finding an appropriate benefit/risk balance.

The impact of adverse effects on their present quality of life was an essential concern reported by most patients. Analgesic adverse effects often affected patients' social, professional and personal living conditions. Patients regretted that some adverse effects interfered with their activities as much as the pain itself. Adverse effects affecting cognitive functions (e.g., memory loss, brain fog), physical aptitudes (e.g., tremors, fatigue) or self-esteem (e.g., weight gain) were particularly concerning for patients. Several patients were concerned that these adverse effects could be detrimental to their social and familial relationships:

One of the side effects is that it completely throws off your sleep cycle, so you no longer keep the same hours as everyone else. That complicates your social life and your family life, because you're totally out of the loop, you can't keep up with the others. (Patient, Part.8-FG2)

Some pharmacists reported being committed to reducing the negative consequences of adverse effects on patients' quality of life, which they mainly appraised through functionality (e.g., ability to work). However, among pharmacists, adverse effects affecting patients' quality of life were usually considered of lower priority than those impairing patients' physical health in the long term. Indeed, some adverse effects were mostly invisible but could lead to long-term serious negative consequences, such as chronic diseases (e.g., kidney dysfunction), incapacities (e.g., because of a fall) or vital risks (e.g., respiratory arrest following an overdose). Pharmacists were particularly mindful of these long-term serious adverse effects that they often found difficult to identify. When they prioritized prevention of irreversible harms over the quality of life, they reported facing some incomprehension from patients:

We pharmacists can sometimes get concerned about things that patients aren't all that worried about, such as kidney function or long-term adverse effects. Patients don't necessarily see that aspect. What bothers patients, I think, is whatever affects their quality of life. (Pharmacist, Part.32-FG8)

Nonetheless, adverse effects presenting risks for future health were also deemed concerning by many patients. The perspective of long-term harms increased patients' uncertainty and powerlessness regarding their prognosis. Many felt they lacked information regarding such serious adverse effects. Some believed that pharmacists, physicians and/or pharmaceutical companies voluntarily hid harmful adverse effects, which highlighted their need to obtain clear information and be involved in their treatment choices:

I've done a lot of searching on the Internet and all I found were things from pharmaceutical companies saying you need to take this medication regardless of everything else, even if I end up ruining my liver. They really downplay the number of people who have serious side effects, like liver cancer, and all kinds of other side effects. (Patient, Part.18-FG4)

3.1.3 The dilemma between pain relief and adverse effect prevention FG discussions suggested discrepancies between pharmacists' and patients' views of the importance given to pain versus potential serious harms. Seeking a balance between medication benefits and risks often entailed choosing between pain relief and adverse effect prevention. Many pharmacists insisted on their challenges in dealing with this ethical dilemma:

I have to take her [an older patient] off one of her pain medications, but then she starts to have pain again because of me. But if I don't remove it, she's going to fall and break her hip. That puts me in an awkward dilemma, it's a choice between having pain or falling and breaking something. (Pharmacist, Part.34-FG9)

Several pharmacists explained that they could opt for dosage lowering or treatment cessation to limit adverse effects, even though the pain would remain unrelieved. For some of them, analgesics were not deemed ‘necessary’ medications because chronic pain was not a life-threatening condition:

Any adverse effect can lead to stopping the treatment. The good thing, if you like, with pain medications, is that the treatment isn't necessary. For other conditions, heart problems, for instance, there are lots of preventive treatments [that are necessary]. When it comes to treating chronic pain, if the medication doesn't suit the patient, given that the purpose was to relieve pain, to improve quality of life, I have no problem taking them off it. (Pharmacist, Part.32-FG8)

Some patients too preferred enduring more pain than facing harmful adverse effects. However, for many patients, the affirmation that chronic pain was not life-threatening was not so evident. Several reported that pain could be so intense that it was hardly tolerable to live with it. They were thus willing to accept intense adverse effects as far as the pain would be effectively relieved:

It's not pleasant, but without them [painkillers], I wouldn't be here still. They're my crutch. I really don't have a choice at the moment. My medication is saving my life in a way, despite these problems of discomfort I have to put up with. I need this medication, regardless of the side effects. (Patient, Part.6-FG2)

Finally, in many cases, the treatment failed to effectively relieve pain in addition to having adverse effects, leaving patients with a negative benefit/risk balance. Some patients questioned the relevance of taking pain medications under such conditions.

3.2 Managing adverse effects in the patient–pharmacist relationship

FG discussions underscored that the management of analgesic adverse effects took place both in patients' daily living activities and during interactions with prescribers and pharmacists, sometimes generating tensions in the decision-making.

3.2.1 Self-management of the pharmacotherapy and adverse effects: Negotiating patients' expertise To reduce adverse effects, several patients reported attempts to self-manage their pharmacotherapy independently from healthcare providers. This included modifying the dosage or the moment of intake of a prescribed drug, as well as self-medicating pain or adverse effects with over-the-counter and/or natural products. Several patients considered they had progressively acquired skills to manage their pharmacotherapy and better control the impact of adverse effects on their daily living. They claimed expertise for making decisions regarding their treatments (e.g., dosage adaptations). However, patients' experiential knowledge and self-management practices were often subject to disagreements from physicians and pharmacists. Prescribers were often described as reluctant to involve patients in decisions regarding pain treatments:

I get told off for not taking my painkillers often enough. I try to discuss with my doctors, because you have to weigh the side effects against what you want to do. Yes, I should take it, but I can't follow the doctors' instructions because of the adverse effects, otherwise, I'd have to change jobs, there'd be more impact on my personal life. (Patient, Part.21-FG5)

Many pharmacists, for their part, described the management of adverse effects as negotiations and patient education, rather than shared decisions. They were concerned by patients' other sources of information on their treatments (e.g., the internet, or patients' relatives). Several of them insisted on the importance of deconstructing patients' ‘beliefs’ regarding medications and tried to discourage patients' initiatives to modify dosages. Pharmacists were often worried about patients' self-medication with natural medicine, as they thought patients often underestimated adverse effects and interactions with other drugs. They found it challenging to obtain information from patients regarding these practices and tried to educate them about potential harms while maintaining the dialogue:

What I explain to them is, let's say a given product contains 10 different natural products, then obviously with those 10 different things, there's definitely going to be some interactions with the 7 medications you're already taking. So, do what you like, but I'm not necessarily recommending that you take it, and if you do take it, and there's some reaction, then come back and see me again. (Pharmacist, Part.39-FG10)

However, some patients described physicians and pharmacists accepting shared decision-making. Having professional experience in the healthcare field was reported as a facilitator to obtain providers' recognizing of their experiential knowledge as patients:

I used to work in health care, and my anaesthetist at the pain clinic leaves me lots of latitude to manage my medication, so I have a major advantage in being able to self-manage it. I'm the one who tells him: ‘Prescribe me more of that’ or, ‘I stopped taking that’. (Patient, Part.8-FG2)

Other patients felt unsatisfied with the self-management of their pharmacotherapy, especially when they perceived it as the result of insufficient involvement of providers rather than their own choice. These participants felt left alone with medication-related risks and adverse effects, as for this patient treated with opioids:

My orthopaedist, when I last saw him, he said, ‘Do you need more?’ So, am I the one who decides? And the pharmacist suggests but doesn't impose anything. So, I'm the one who's managing my medication. I could have far more than I'm getting at the moment, but I think I'm already taking quite a lot. There's no one putting the brakes on for me. I'm the one making the decision, it's a little sad, but that's the way it is. (Patient, Part.6-FG2)

3.2.2 Managing opioid adverse effects: Pharmacist's policing practices In the case of opioid treatments, pharmacists reported that their involvement in the management of adverse effects often took the form of monitoring and policing patients' medication use. Several patients using opioids for their chronic pain were concerned that pharmacists could suspect them of nonmedical use, and threaten the continuity of their treatment:

The first thing she [my new pharmacist] did was to report me to the College of Pharmacists, claiming I was a drug dealer. In her mind, it was impossible for someone to take so many drugs and still be able to walk and talk. (Patient, Part.23-FG6)

Several pharmacists believed that the ‘police officer’ role was part of their missions with patients using opioids for chronic pain, though some of them disliked these tasks that were at the boundaries of healthcare:

As far as dependency goes and managing opioids, if you say to them, ‘No, I can't serve you today. You've still got some left at home’, then the person comes up with excuses. So, you might believe them the first time, but at some point, it's no longer acceptable. You have to set dates, to demand that people bring their medications into the pharmacy to see what they have left. All that kind of thing is really no fun. All that policing side of it, that can be really annoying sometimes. (Pharmacist, Part.41-FG11)

Furthermore, opioid overdose prevention could cause challenging dialogue between patients and pharmacists. Naloxone, for example, was seen by several pharmacists as an important tool to reverse the effects of overdoses. However, patients facing opioid-related stigma in their daily life were not likely to accept naloxone because they felt negatively labelled as opioid-dependent or recreational consumers:

We pharmacists have the means to give out a kind of opioid EpiPen® called naloxone, so that patients can have it at home if they overdose on their opioids. So, if, for some reason, because they were in pain, they've taken a little more than they should have, and they have a breathing arrest, they can take naloxone. But getting patients to give this solution a try is a whole new challenge, because when you suggest it, they say, ‘I'm not a drug addict. I take medication to relieve my pain, but I'm not a junkie living on the street, so I don't think I need that’. It's really stigmatized. (Pharmacist, Part.38-FG10)

3.2.3 Healthcare and business running: Tension in the pharmacist's role Many pharmacists considered they had, or should have, a key role to play in the management of analgesic adverse effects, including patient follow-up, treatment optimization, patient education and support:

It's really important to follow up with the patient because some side effects are temporary and sometimes, just getting a bit of encouragement to keep going will help the patient to continue long enough to see the benefits. (Pharmacist, Part.30-FG7)