A Crack in the Wall: Chronic Pain Management in Integrative Group Medical Visits

Abstract

Amidst a national crisis of opioid overdose, substantial uncertainty remains over how to safely and effectively address chronic pain. In response to this crisis, safety-net primary care clinics are instituting integrative group medical visits (IGMVs) for chronic pain management. Through two qualitative studies of IGMVs, we found that these groups acted as workarounds implemented by clinicians seeking to innovate upon standard pain management protocols. While clinical uncertainty is often framed as a problem to be managed, in this instance, overlapping uncertainties provided an opportunity through which enterprising clinicians could generate reform at the local level. However, these clinician-led changes were incremental, situational, and partial, and occurred outside of broader systemic reform. In the following article, we draw on 46 interviews with clinicians and staff associated with IGMVs and observations of 34 sessions of 22 distinct IGMVs. We begin by describing the structure of the IGMVs we observed. We analyze the multiple uncertainties surrounding chronic pain and its treatment at the time of our data collection, just before the opioid crisis was declared a national public health emergency. We then demonstrate how clinicians tinkered with existing pain management protocols via their involvement with IGMVs. Lastly, we discuss the conditions of possibility that allowed for the existence of IGMVs at our study sites, as well as the conditions of limitation that restricted the expansion of these groups. Our research points to the potential of IGMVs for treating chronic pain, while showing that IGMVs continue as an innovation by individual clinicians, not as a result of broader reforms.

Introduction

“You think we’re trying to take down the whole wall? It’s enough to make a crack …[i]n order to imagine everything that could be done tomorrow.”

–S.I. Galeano, 2015

Chronic pain is a condition marked by uncertainty, given its subjective nature and its capacity to evade language (Scarry, 1987). Recently, pain has been rendered all the more uncertain in light of the opioid-related overdose crisis, which has left clinicians unclear about how to proceed with pain management strategies given the risks associated with these once ubiquitous drugs (Knight et al., 2017). Primary care providers (PCPs) are under increasing pressure to reduce or discontinue their patients’ use of opioid pain medications, yet often cannot offer their patients accessible and affordable alternatives (Dowell et al., 2016; Hurstak et al., 2017).

In this paper, we demonstrate how clinicians working in safety net settings responded to the multiple uncertainties surrounded chronic pain management by tinkering with standards of care, creating workarounds at the local level. These workarounds were made possible by certain conditions of possibility, and their capacity to expand was constrained by particular conditions of limitation. Thus, uncertainties surrounding pain and its treatment created an opportunity through which enterprising clinicians could achieve incremental, local practice changes. .

Chronic Pain Management and Group Medical Visits

Our research focuses on safety-net primary care clinics in the US that predominantly care for patients who are uninsured or publicly insured through Medicaid (National Association of Community Health Centers, 2020). Treating pain in safety-net settings is particularly challenging, given the limited resources of safety-net institutions, patients’ high rates of comorbid conditions, and experiences of individual and community trauma that may cause or exacerbate chronic pain (Pinderhughes et al., 2016). Socioeconomic and racial/ethnic inequities in chronic pain prevalence and treatment mirror those present in other health conditions (Janevic et al., 2017; Meghani et al., 2012a). Race/ethnicity, socioeconomic status, gender and language heavily shape experiences of chronic pain diagnosis and treatment (Hollingshead et al., 2016; Pryma, 2017), while structural vulnerability (Quesada et al., 2011) heightens the uncertainties of living with and seeking treatment for chronic pain.

Patient-clinician relationships in chronic pain management are often contentious, and clinicians struggle with assessing and treating chronic pain (Esquibel and Borkan, 2014), particularly among patients with co-occurring substance use or mental health conditions (Chang et al., 2016). Prior research has shown that individual primary care visits generally provide inadequate time to assess pain, contributing to these challenges (Satterwhite et al., 2019).

Major US medical associations and government agencies including the Joint Commission and federal Department of Health and Human Services have called for reforms in pain treatment (Joint Commission, 2017; US Department of Health and Human Services, 2019), including increasing the use of non-pharmacologic treatments and harm-reduction interventions (Vashishtha et al., 2017). Recommendations for non-pharmacologic treatment often include complementary and integrative health (CIH) approaches like acupuncture and massage (Chou et al., 2017; Nahin et al., 2016), as well as physical therapy and mental health care. However, access to these “gold standard” approaches to pain management is limited, in part because of lack of insurance reimbursement (Meghani et al., 2012b). In particular, access to CIH is stratified; treatments such as acupuncture and massage are rarely reimbursed by insurance, and high out-of-pocket costs makes them largely inaccessible to people who are uninsured or publicly insured (Steyer et al., 2002; Whedon et al., 2017). Despite high levels of interest in CIH on the part of safety-net clinicians and patients (Hurstak and Kushel, 2016), and regardless of national efforts to transition pain management towards nonpharmacological interventions, safety-net patients often cannot access CIH and other elements of “gold-standard” pain care. Though Medicaid programs in several states now offer pilot programs covering acupuncture, massage, and other forms of CIH for chronic pain, and Medicare has initiated coverage of acupuncture specifically for low back pain, such programs were not present in the states where we conducted this research (Clemans-Cope et al., 2017; Davis et al., 2018; Donovan et al., 2017).

In response to these challenges, some safety-net primary care clinics have begun implementing group medical visits that provide integrative health care. These models aim to ameliorate healthcare disparities through a combination of CIH, peer support and education, and standard primary care. Group medical visits (GMVs) are now widespread in U.S. primary care. Commonly used for prenatal care (Byerley and Haas, 2017) and diabetes (Burke and O’Grady, 2012), GMVs bring multiple patients to the same space for medical care, health education, and peer support. GMVs for chronic pain and opioid use disorder described in research literature aim to encourage safer use of opioid medication (Romanelli et al., 2017; Spelman et al., 2017), increase access to medication-assisted treatment such as buprenorphine (Sokol et al., 2018), and support patient self-management (Gaynor et al., 2007). Integrative group medical visits (IGMVs) have primarily been implemented in safety-net settings and expand CIH access for low-income people. Health outcomes research on IGMVs for chronic pain has demonstrated a variety of benefits, including improvements in sleep quality and health-related quality of life, as well as decreased levels of pain intensity, opioid medication use, depression symptoms and fatigue (Chao et al., 2015; Cornelio-Flores et al., 2017; Gardiner et al., 2019, 2014; Mehl-Madrona et al., 2016).

This paper seeks to understand how IGMVs for chronic pain operate in safety-net primary care clinics amidst numerous uncertainties. Our analysis draws upon two qualitative studies of IGMVs for chronic pain and argues that the uncertainties associated with pain in the midst of an opioid-related overdose crisis created a “crack in the wall” (Galeano, 2015) through which safety-net clinicians could tinker with standard pain management protocols. These IMGVs act as workarounds achieved by enterprising clinicians capitalizing on the overlapping uncertainties related to pain in order to generate reform at the clinic level. Clinical uncertainty, in this instance, can be understood as an opportunity, rather than a problem to be managed. However, these clinician-led changes were incremental, situational, and partial, and occurred without broader systemic reform.

Clinical Uncertainty

Pain is characterized by a particular kind of medical uncertainty, given that it lacks a diagnostic biomarker and thus leaves patients needing to relay their discomfort to clinicians using words alone. However, placing pain in a special category of uncertainty does not diminish the uncertainty that surrounds all clinical conditions. Medical sociologists have been finding frames to make sense of the clinical uncertainty that pervades medicine for decades, beginning when Renee Fox (1957) characterized medical knowledge as intrinsically uncertain. She contended that the expanding body of clinical information would produce a proportionately expanding number of gaps in clinical comprehension, and that medical education should prepare clinicians-in-training to face such uncertainties with confidence. Timmermans & Angell (2001) updated the concept of clinical uncertainty after evidence-based medicine came to dominate medical logic, emphasizing the continuing ubiquity of uncertainty despite increased standardization of medical knowledge. These and other accounts of clinical uncertainty (Anspach, 1997; Rafalovich, 2005) have framed ambiguity as something to be managed, circumscribed, or overcome.

Recent scholarship, however, has suggested that uncertainty in the clinical context can also be conceived of as productive (Reed et al., 2016), by creating space for opportunity, tinkering, and resistance through the development of workarounds. As Pickersgill points out in his account of how clinical psychologists serve as gatekeepers to mental health services, “working towards a resolution of ambiguity can involve the imagining of different futures and the animation of affects and action to realise or avoid them” (Pickersgill, 2019, p. 2). Similarly, Moreira and colleagues (2009) demonstrate how the lack of a biomarker for diagnosing Alzheimer’s can open up space in which uncertainty can be leveraged as stakeholders redeploy ambiguity in the “collective production” of new standards of care.

Through our analysis of IMGVs for chronic pain management, we found that uncertainties surrounding the treatment of pain in the midst of the opioid crisis created similar space for the deployment of emerging forms of clinical practice, including IMGVs for chronic pain. When faced with uncertainties about what constituted safe and effective pain management, the safety-net clinicians in our parallel studies responded by creating IGMVs as workarounds. Sociological literature has typically defined workarounds as makeshift solutions developed when standard protocol is perceived to be ineffective (Bowker and Star, 2000; Whooley, 2010). In this case, standard protocol for pain management was called into question via the opioid-related overdose crisis, and clinicians responded by fashioning alternative protocols that aligned with their goals for patient care. While such workarounds can represent a form of resistance, they can also be likened to the practice of ‘tinkering’ (Knorr-Cetina, 1979; Mol, 2008; Van Natta et al., 2018) in which solutions are crafted to meet immediate and localized needs.

Methods

This article draws on data from two grounded theory projects conducted by different researchers with overlapping timelines in 2016 and 2017. While each project was designed independently , findings regarding IGMVs were similar enough to warrant this joint analysis. We draw from types of data that were available in both projects, specifically interviews with clinic staff and ethnographic observations of IGMVs for chronic pain.

Setting:

Our research was conducted at six public or non-profit safety-net health care organizations, four in the San Francisco Bay Area and two in the Boston area. In some cases, we observed IGMVs at multiple sites within a single organization, for a total of 11 clinic sites. All served racially and ethnically diverse, predominantly low-income, multilingual patient populations (specific demographics are not shared here to protect anonymity of sites). We purposively sampled for clinic sites that offered IGMVs, defined as including billable biomedical care, one or more forms of CIH in most sessions, peer support, and health education. Since they included medical care reimbursed by insurance, all IGMVs were facilitated by a licensed clinician (typically physician or nurse-practitioner). Most IGMVs were co-facilitated by another staff member, including psychologists, health educators and medical assistants, among others. Forms of CIH offered in IGMVs included acupuncture, meditation and other mindfulness practices, yoga, massage, and osteopathic manual medicine. In this article we use the term “clinician” to refer to licensed health care providers (e.g. physicians, psychologists, acupuncturists), and “staff” to refer to healthcare workers more broadly (including medical assistants, health educators, etc.).

Data collection:

In total, we conducted 114 hours of ethnographic observations of IGMVs at 11 safety-net clinic sites. Observations were conducted by the authors after obtaining verbal consent from all present parties, and focused on the structures and interactions occurring within each group. In addition, we interviewed 43 clinicians and other staff, either in person or by phone, about their pain management practices; 30 of the 43 interviewees had experience coordinating and/or providing care in IGMVs.

The first project (ATL) was a mixed-methods study of IGMVs in safety-net clinics. For this project twenty-eight semi-structured interviews were completed with physicians, health educators, and other staff; 23 of these were involved with IGMVs focused on chronic pain. Staff interview questions focused on experience providing care in IGMVs, perspectives on CIH, perceived effects of IGMV participation on patients and staff, and organizational support for IGMVs. ATL observed 20 distinct IGMVs, 15 of which primarily served patients with chronic pain. Patients and staff provided verbal consent at the time of observation; observations were recorded in detailed field notes. While observing IGMVs, I invited patients to participate in individual interviews. Eligibility criteria included age 18 or older and fluency in English or Spanish. I interviewed 25 patients by phone or in person; 18 of them reported chronic pain. Demographic data was collected through brief questionnaires. Interviews and observations were conducted in English and Spanish and professionally transcribed. Chronic pain and the use of opioid medications were emergent themes in a project that was not designed to focus on a particular health condition.

The second project (SR) was an ethnographic study examining the impacts of the opioid crisis on the clinical treatment of chronic pain. As part of this study, SR conducted longitudinal observations of 14 sessions of two IGMVs for chronic pain. Due to agreements with Institutional Review Boards, fieldnotes focused exclusively on the content of the curriculum and staff activities, and excluded any information about patients and their health. All group members gave verbal consent to observations, understanding the content of fieldnotes would not include patient-related details. The 15 clinicians interviewed for this study worked within 6 different institutions and participated in semi-structured interviews with SR . Questions focused on standards of practice in pain management, perceived safety and efficacy of opioids, perspectives on appropriate CIH modalities for safety net patients, and professional backgrounds. Eight clinicians were involved in facilitating pain groups. Clinicians’ roles included physicians, nurse practitioners, physical therapists, licensed social workers, and psychologists. Several were engaged in clinical research, and some taught medical students and residents. Eligibility for interviews included actively practicing medicine, working with a chronic pain patient population, and proficiency in English.

Data analysis:

Both projects employed constructivist grounded theory methods, including iterative adjustments to data collection and analysis, coding and ongoing memo writing about data (Charmaz, 2006; Clarke, 2005). Field notes and interviews from each project were coded separately. The original concept for this paper was developed by ATL based on data from her project. After data collection was completed, we analysed overlapping codes from the two datasets side-by-side in order to develop an understanding of the roles that IGMVs played in safety net clinics. These overlapping codes related to benefits of group visits for chronic pain, integrative approaches to chronic pain, trauma and chronic pain, and changes in opioid prescribing guidelines and practices. We drew on data in these areas from both studies to develop this manuscript. These studies were both approved by the UCSF IRB (Study #15-18421 and #16-19919]. We use a combination of pseudonyms and real first names, according to study participants’ anonymity preferences.

Results

These findings draw from ethnographic observation of 34 sessions of 22 distinct GMVs as well as interviews with 46 clinic staff involved with GMVs. Forty-six percent of these were licensed clinicians, while the rest worked in roles including health educators, medical assistants, and program managers. Seventy-six percent of interviewees identified as female. We found that both chronic pain and its management created an environment of clinical uncertainty in the safety-net clinics where we conducted our studies. Clinicians responded to these uncertainties by creating workarounds in the form of IGMVs, which address multiple elements of physical and mental health. Particular conditions of possibility had to be in place in order for these programs to emerge, and several conditions of limitation resulted in constraints on the adoption and expansion of IGMVs. Through ethnographic observation and interviews with IGMV staff, we found that IGMVs have taken shape through tinkering with standard care, and represent a “crack in the wall” of organizational practices, rather than a major overhaul of how organizations deliver care.

What does the crack in the wall look like?

When patients walk into an integrative group medical visit, they typically bypass the waiting room and enter a large room where they check their blood pressure, weight, and other vital signs, sometimes with assistance from other patient or staff members. Vital signs are collected to inform clinical decision-making and meet insurance billing requirements. Most IGMVs ask patients to rate their pain numerically from 0 to 10 on a standardized pain scale alongside other vital signs. In one group, I (ATL) sat in a circle with patients, a clinician, and a staff co-facilitator and listened to patients share their current scores on the pain scale as well as the “joy scale” developed by Charles, a patient in the group. That day, Charles said his pain was a 3 and his joy was an 8. Another patient said her pain was 10/10—"okay, make that 8 out of 10” --in her back, hip, and knee. As for the joy scale, she was at a 5 because she was “fighting with hospice, sisters, attorney, step kids, great grandkids.” Later, Charles explained that he got frustrated that people always asked about his pain. “How come nobody ever asks me how my joy level is?” In our interview, Charles shared that he had used opioids to treat a painful degenerative condition for decades before joining IGMVs. After a year of weekly IGMV attendance, he had substantially decreased his use of opioids and other medications. Charles changed the way care was provided in his IGMV by sharing critiques of the pain scale and suggesting other ways for patients to describe fluctuations in their health. Two years later, one of these clinicians told me, “Now we consider the joy scale one of our vital signs [in group visits], a way to assess wellbeing.” Here, the IGMV format created space for a patient’s innovation; support from clinicians and the presence of peers meant this innovation was taken up by other patients. The joy scale, like the pain scale, quantified health, and the familiar format may be why clinicians readily accepted this innovation. This is one of many examples of IGMV patients feeling seen and heard by clinicians, who in turn described being more satisfied with the care they were able to provide in IGMVs than in individual care.

Safety-net clinicians responded to uncertainty, limited resources and lack of public investment in gold-standard pain care by developing IGMVs: a multimodal form of treatment that they believed met patient and clinician needs alike better than standard individual pain management. They responded to pain treatment guidelines that were not necessarily written with their patients in mind and innovated with available resources in collaboration with their structurally vulnerable patients. All IGMV models included medical care that could be reimbursed by Medicaid and other insurance using the same billing practices as individual primary care visits (American Academy of Family Practice, n.d.), though the CIH care provided was not directly reimbursable. These insurance reimbursements were the primary form of funding for IGMV programs; though several were supplemented by research or philanthropic grants, the majority were funded out of general clinic budgets, and expenses were generally limited to staff time and snacks for patients.

We found that IGMV clinicians responded to uncertainty about opioids in several ways. Though we did not systematically ask about opioid use among IGMV patients, it was clear from observation that patients with a wide range of pain-related diagnoses and experiences with opioid medication participated in groups. In some groups, clinicians prescribed opioids and mandated urine drug tests. Other IGMVs required individual visits with patients’ PCPs for opioid prescriptions. Though opioids were not the primary focus of any program, many IGMV clinicians were oriented towards reducing opioid use and eliminating it where possible.

IGMVs were structured in a variety of different ways. We observed some groups with structured curricula, such as those modeled on Mindfulness-Based Stress Reduction or Cognitive Behavioral Therapy. Others focused on particular forms of CIH such as yoga or acupuncture, while still others invited patients to shape the group’s content. Noemi, a staff member in multiple, less-structured IGMVs, described how sessions included “activities with the groups…whether it's cooking, meditation, writing a book. It just depends on what the groups want to do.” Given the lack of funding for CIH, clinicians generally made decisions about what kinds of CIH to offer by drawing on existing resources at their workplaces. For example, one clinic had staff trained in acupuncture and massage, and offered weekly, ongoing acupuncture and massage GMVs to a rotating set of patients. Another clinic developed a structured curriculum based on Mindfulness-Based Stress Reduction (Kabat-Zinn, 1990) and provided a nine-session group to a consistent set of patients. These varied staffing configurations depended primarily on the training of existing staff members, who typically played multiple roles at their workplaces (e.g. a licensed massage therapist who primarily worked as a coordinator of clinical programs, a licensed acupuncturist who primarily worked in an administrative role). In many cases, staff provided CIH exclusively in group visits because the lack of reimbursement meant it was not feasible to offer in individual visits. Working with existing resources made access possible. Where individual primary visits typically last 10-15 minutes, GMVs are generally between one and three hours long, providing adequate time for this blend of activities. Katherine, a clinician, explained why she developed IGMVs for chronic pain:

Working in a very busy, underserved setting…there is a lot of depression, and there's a lot of pain. Those are two really difficult needles to move in the setting of traditional 15-minute visits. We were also…really wanting to see how we could drive some [integrative] services into what we do. It seemed very obvious to bring together this mindfulness piece with the integrative medicine [and] the medical group visit.

Katherine, like other clinicians, began IGMVs to bring CIH into primary care and applied it to chronic pain and depression because these were challenging to treat in individual care.

Overlapping Uncertainties

Chronic pain and clinical uncertainty about its treatment has been a long-standing challenge in primary care. Chronic pain is a condition without a biomarker, meaning that there is no biological test available to verify its veracity (Rubin, 2018). The uncertainty surrounding pain and the associated difficulties with arriving at a diagnosis were often described as difficult for both clinicians and patients. One clinician discussed the challenges that her patients faced when navigating the uncertainty of pain:

Even how severe pain is, the idea that, "Oh, something will get better," or even knowing the "why" you're having pain. A lot of our patients have pain, don't have a diagnosis for why they have pain or have some idea but feel like they don't have a definitive answer on what's going on. And that makes it so much harder to accept it and manage it…Because then there's always that thought of well, did somebody miss something or is this--Will they figure out eventually and it be cured? It's hard to then have any acceptance of it if there's that uncertainty.

In this clinician’s assessment, lack of certainty often results in a lack of acceptance, and therefore increased difficulty with self-management of pain. While all medical conditions are marked with some degree of uncertainty, this clinician’s account demonstrates that pain occupies a special position due to its ability to evade definitive diagnoses. The lack of a biomarker also necessitates that clinicians rely on patients’ self-report of experienced pain, which can place strain on both parties and can serve as an entry point for stratifying processes (Netherland and Hansen, 2017; Rubin, 2018).

In contrast, some clinicians spoke of the ontological flexibility of pain as a point of entry for new ways to think about and, by extension, intervene upon symptoms. One referring clinician described the benefits of her clinic’s IGMV, pointing to the emergence of multimodal approaches to pain management:

There are so many factors that actually affect whether or not someone has an experience with pain.… That’s part of the reason why we have a multimodal treatment now. Because we know that there are various part of the nervous system that we can impact in an effort to try [to] improve a person's pain experience. [Talking about that with patients is] actually difficult to do in a short visit…If you sat me down in a room with a patient one-on-one, it's very hard to have a well-packaged short spiel. It's not something that's easily heard or digested. Especially by someone who's suffered with significant pain.

In this clinician’s estimation, understanding pain as something that derives from the nervous system motivated the implementation of multimodal treatment, in this case IGMVs. The numerous factors influencing the experience of pain were not framed as a difficulty, but were rather seen as leading to innovations in treatment. Her clinic’s IGMV focused on “psychoneuroeducation,” i.e. educating patients about how pain operates in the nervous system. However, explaining this technical information to patients was seen as challenging in a brief individual visit, the only kind of care available to them outside of the IGMV. For clinicians seeking a multimodal approach, rather than simply prescribing opioids, IGMVs allowed more space and time to convey the complexities of pain and its treatment. These clinicians saw the complexities and ambiguities of pain as necessitating a different treatment model.

Many clinicians we interviewed concurred that they found it very challenging to treat chronic pain in the brief individual visits that make up standard primary care. Laura, a clinician who had developed chronic pain IGMVs, explained that “chronic pain is the bane of [clinicians’] existence in some ways, but [group visits are] actually a really fun way to work with chronic pain patients.” Where individual visits for chronic pain were often fraught for both patients and staff, IGMVs provided space for the emotional components of pain as well as immediate experiences of pain relief through CIH. Adisa, a clinician, explained this as her motivation for providing IGMVs, saying, “I wanted to do something about chronic pain because people were always coming back to us [in individual visits] and nothing was getting better.” Given the challenges of individual chronic pain visits, some clinicians initially assumed that treating multiple patients with chronic pain together in an IGMV would be even more difficult. Instead, many found groups to be a better experience for all involved for a variety of reasons, including more time and access to both CIH and peer support.

At the time of our research, the challenge of treating pain in primary care had increased as the opioid crisis raised the stakes of pain treatment. Recent changes in understandings about the effectiveness and safety of opioid medication increased uncertainty about chronic pain treatment. One clinician described her confusion in light of these changes:

There was a time, for example, when we were told, "Use long-acting opioids because that's safer and better than using multiple short-acting ones." And so that's actually your ideal is to use something like long-acting morphine, or methadone, or something like that. And then the most recent CDC pain guideline said, "Short-acting is preferred." So, it can be very confusing.

Changing federal guidelines about how to safely and effectively treat pain using opioid medications resulted in confusion for this clinician and many others. These changes frequently left clinicians feeling as though they did not have safe alternatives for their chronic pain patients, as described by a clinician who had been working in primary care during the 1990s when opioids were promoted as safe and effective:

Patients often are seeking opioid medications in quantities or types that we no longer want to prescribe. That sea change [in how we think about opioids] has occurred and… I don't know, some of the boats are getting stranded…we don't have particularly good tools. Again, my highest goal is the safety for patients.

This clinician identified a “sea change,” or pendulum swing in opioid logic, that characterized shifting understandings regarding the safety and efficacy of opioids. Now that federal guidelines promote tapering chronic pain patients off of long-term opioids, some clinicians are left feeling as though they have few options for treating their patients’ pain. Amidst these changes in how opioids are prescribed, he pointed to the “boats [that] are getting stranded,” referring to patients who were left without accessible treatment because of changing policy around pain management. Safety-net clinicians were responding to uncertainties surrounding pain itself as well as its treatment in the midst of an opioid-related overdose crisis.

Tinkering and Workarounds

The ambiguity of pain and the risks associated with opioids have resulted in a growing interest in non-pharmacologic pain treatment, but resources to increase access to gold standard care have often been unavailable in safety-net settings. Laura described the process of developing her IGMV programs:

Believe me, the program was just patched together.… [Working here] is like standing in front of a stone wall and you have to just be scanning and looking for a crack and when you see a crack you just have to make like water and you have to go through and create a little something.

Laura’s metaphor to describe the process of developing IGMV programs within her institution parallels how staff at other organizations spoke about their efforts. IGMVs emerged as a safe alternative or complement to opioids, but one that required staff to work around existing organizational practices and tinker with available resources.

IGMV programs were made possible in part because both patients and clinicians were open to CIH, an openness stemming in part from growing concern about opioid safety. Jossalyn worked in a clinic that primarily served low-income African-American patients and was actively limiting opioid prescribing in response to safety concerns. Her clinic had determined that IGMVs were a pragmatic way to expand access to CIH and offered multiple IGMVs that served patients with chronic pain. Jossalyn explained:

[Many patients] feel that their pain is at a level where they'll try anything. I'm going to take all these pills and it’s not working. I've tried physical therapy [and] it’s not working. What else is there? The primary care providers give [the IGMV programs] the best recommendation because the doctor’s like ‘okay, we've tried all this stuff, you really need to try acupuncture. This will really alleviate your pain…’ then we call and say "hey, you know I got this referral from doctor so and so" and [the patient says] ‘oh, yes great when can I get in [to the IGMV].’ Acupuncture changes people's lives here.

Aware of the continuing challenges around access to CIH, we sought out clinics that serve low-income people from a range of racial/ethnic backgrounds and found that many were interested in CIH when given access. Jossalyn underscored that while the effects of IGMV participation on her patients’ health was notable, their willingness to participate should not be surprising given the lack of other effective and accessible treatment.

IGMV clinicians responded to uncertainty about opioids in a variety of ways, with most emphasizing principles of harm-reduction. Some clinicians encouraged patients to tinker with daily habits that might affect their pain, perhaps smoking a bit less and meditating a bit more. Clarissa, an IGMV staff member, said of her IGMV’s approach: “We're not purists. A lot of people smoke cigarettes, smoke pot, eat a lot of junk. We never say, "stop doing that," we just say, "start doing a little more of this, this and this.” Rather than mandating that patients discontinue opioid medications, many IGMV staff instead encouraged patients to incorporate CIH into their self-care practices and draw on the support of the group. Clarissa’s colleague, Tara, said they hoped that:

[Patients’] reported pain is going to decrease, so their experience with their pain will decrease and their depression will decrease. So far, we've seen that…A big part of it is not changing the pain and that's the mindfulness part…We're not asking you to go off of your medication for these things. The hope is that they'll develop a different relationship with their pain, and I've definitely seen that.

Part of tinkering with health-related behaviors included encouraging patients to think about pain differently—rather than expecting it to go away completely--and using CIH practices such as mindfulness meditation or yoga to make this possible.

Conditions of Possibility and Conditions of Limitation

We found several conditions evident at all sites that made it possible to implement IGMVs for chronic pain in safety-net clinics, the most notable being the presence of clinicians with particular commitments to IGMVs for pain management. At the same time, various constraints limited the adoption and expansion of IGMVs in safety net clinics. These limitations resulted in additional uncertainty about who could benefit most from particular kinds of treatment, as well as continued stratification of access to care.

The presence of clinicians who were personally committed to group medical visits and/or CIH enabled IGMVs to proliferate. The development of IGMVs was not driven by policy change or patient requests. Instead, motivated clinicians with particular backgrounds made specific kinds of programs feasible and in some cases pushed for organization-wide IGMV expansion. Some clinicians who had personal experience with chronic pain were particularly motivated to implement IGMVs for pain management. Laura said, “I deal with chronic back pain so I also felt…that chronic support [is] just needed for chronic illness.” This idea of “chronic support” led her to develop a drop-in IGMV program for chronic pain, so that patients could participate with whatever frequency and duration they found most helpful. At another site, a staff member had witnessed the intersection between pain and trauma within her own life and pushed for the creation of an IGMV that addressed pain and PTSD together. She spent over six years convincing clinic administration that it was worthwhile. She said she eventually succeeded through:

…telling them about trauma and pain…and that what I really wanted to do was give people tools to lower their opioids, and it happened to coincide with the pendulum swinging…It was also at the same time where integration is the name of the game, and so, we're supposed to do integration and this is still more integrated than people are ready for, I think.

She named various structural conditions that allowed for the establishment of her IGMV, pointing to the swinging of the pendulum in opioid logic and an increased interest in integrated physical and mental health care. However, she concluded by referencing her concern that the group she facilitated was “still more integrated than people are ready for,” foreshadowing the fact that her administration cut funding for her IGMV shortly after our interview.

One condition limiting the expansion and effectiveness of IGMVs for chronic pain was uncertainty about how to provide high-quality CIH care with limited resources. While clinical uncertainties about optimal dose and duration of particular CIH treatments do exist, bureaucratic uncertainty played a larger role in limiting safety-net patients’ access to CIH. For example, Louisa’s clinic administration allowed her to provide acupuncture in IGMVs once a month at maximum because she had another primary role in the clinic and acupuncture was not reimbursed by patients’ insurance. However, her clinical perspective was that people with longstanding chronic pain should receive ongoing weekly acupuncture for maximal benefit: “[Monthly acupuncture in] group care is super great if and when you can find space for that stuff. It's not frequent enough to be considered an actual treatment [for chronic pain].” Acupuncture in her IGMV, then, acted as an introduction to a potentially helpful treatment that was not always accessible. Clinics made do with available resources, developing systems of referral and scheduling that maximized how many patients could receive some treatment while still unable to serve all patients who could have benefitted.

The absence of frequent, long-term CIH treatment despite clinician and patient interest was a direct result of the fact that Medicaid and Medicare did not generally reimburse for CIH, even approaches that were recommended by major medical organizations. Lingering uncertainty about what high-quality, CIH-inclusive pain management entails, and whether low-income people deserve the public investment required to provide it, has resulted in reimbursement policies limiting the accessibility of CIH. This had explicit effects on how programs were structured, and some staff members argued that IGMVs could have been more effective with more consistent access to practitioners with extensive training in CIH (e.g. licensed acupuncturists). IGMVs acted as a workaround, in that the multimodal approach allowed for particular kinds of CIHeven when directly billing insurance for these services was not possible .

Although IGMVs were developed in part to address stratification in CIH access, some level of stratification continued in these programs. This appeared to be in part because of their emergent, opportunistic structure. All patients and clinicians agreed that individual patients should be able to choose whether to participate in IGMVs. However, substantial disagreement remained over which patients should be invited to participate. Stratification was most visible in limited access to IGMVs for non-English speakers and people with severe mental health conditions, with notable exceptions.

For example, the majority of our research sites only offered IGMVs for pain in English despite offering other GMVs in Spanish. When asked about the absence of Spanish-language IGMVs for pain, one staff member noted that “There haven't been any clinicians who have expressed interest in… taking on a [Spanish] group.” In addition to underscoring the role of individual clinicians in determining IGMV access, she and others hypothesized that fewer Spanish-speaking patients used opioids, though none pointed to local data to confirm this. At no site was there discussion of offering chronic pain IGMVs in languages other than English or Spanish or providing interpretation in IGMVs, though most clinics served substantial numbers of patients who spoke other languages.

According to IGMV staff , most groups included people with active addiction, concurrent mental health diagnoses and extensive experiences of trauma. However, some sites restricted IGMV participation to patients who were not already receiving mental health services, preventing many patients with mental health diagnoses from participating. At other sites, PCP s and IGMV facilitators made informal decisions about who would be an appropriate IGMV patient. We observed patients sharing traumatic experiences with other group members, and heard from patients and clinicians about how IGMV participation could be both beneficial and challenging for patients with past or ongoing trauma. In practice, all groups included patients with trauma experiences.

Individual clinicians’ comfort with particular mental health conditions shaped which patients were welcomed into IGMVs. Bethany explained that at her organization, “there's an unspoken, or informal [understanding]…if somebody's moderately to severely mentally ill, they shouldn't be in group. Basically, that's about disruption, disruptive behavior, not about them.” This perspective was echoed by other clinicians. In contrast, Adisa described how she supported patients with mental health or neurological issues to participate in IGMVs successfully:

People over-medicated on their schizophrenia [medication], where they fall asleep in their group… I'd say, ‘Mr. So-and-so, I think we better talk to your [primary care] provider about your Haldol’… We get it adjusted, but I haven't had anything where it's gotten uncomfortable, rowdy, violent.

Adisa saw some patients’ struggles to participate in IGMVs as an opportunity to improve their care rather than exclude them.

Disagreement existed across (and, at times, within) sites about whether IGMVs should include patients with extensive experiences of trauma. Some sites had an explicit policy excluding patients with diagnosed PTSD or ongoing trauma exposure, seeing them as disruptive to group cohesion. Other clinicians described IGMVs as particularly beneficial to traumatized patients. As one clinician shared:

If I had my way, I would have every single chronic pain patient in a group.… It allows people's voices to be heard and people to be understood and people to share, to commiserate.… Chronic pain is incredibly isolating.… Having people be together automatically makes it better.

This clinician emphasized how shared experience among patients helped chronic pain by breaking isolation. She made the case that by not limiting the program to patients who were already coping well and adhering to treatment recommendations, IGMVs provided more meaningful support. Such variation existed in part because IGMVs were developed and sustained by individual clinicians; if these programs were implemented at an organizational level, attention to stratification might be greater. However, expansion of IGMVs also could result in diagnosis-based referral guidelines that might exclude more patients with mental health diagnoses.

Discussion and Conclusion

In this article, we argue that IGMVs are an emerging intervention made possible by a set of uncertainties related to pain, opioids, and CIH, intensified by safety-net patients’ limited access to recommended multimodal care. Enterprising clinicians responded to these overlapping uncertainties by tinkering with standards of care, creating workarounds in the form of IGMVs for chronic pain. We have also demonstrated how expansion of these programs is limited by lack of insurance reimbursement, related to stratified access to care and continued uncertainty about optimal pain management.

The ongoing opioid crisis has created space for new kinds of healthcare. This parallels the early years of HIV/AIDS, when the absence of effective biomedical treatment resulted in some openness to CIH as supportive care. For example, federal Ryan White funds were used in San Francisco and other locations to make acupuncture and other treatments accessible to people with HIV/AIDS (Moffett et al., 1994). Chronic pain treatment recommendations increasingly include CIH, but access to treatment is limited for structurally vulnerable patients. Safety-net clinicians demonstrate the potential of IGMVs to increase access to multimodal care. Safety-net clinics have a long history of taking up innovative approaches to care (Boehmer et al., 2018; Geiger, 2016). IGMVs are a newer innovation, alongside efforts to reduce opioid prescribing and increase the use of medication-assisted treatment for opioid use disorders (Doorley et al., 2017; Suzuki et al., 2015)

Our research points to the potential of IGMVs for treating chronic pain and decreasing opioid use. Yet IGMVs continue as an innovation by individual clinicians, not as broader reforms. Such ad hoc innovations in clinical practice are both flexible and limited in what they can achieve. Low-income people’s limited access to CIH helped make IGMVs possible, even as lack of reimbursement limited their growth and scope. Lack of reimbursement for multi-modal pain care and CIH more broadly has consistently limited access to these forms of care (Meghani et al., 2012b), yet attention to their benefits are increasing in response to concerns about opioid safety (US Department of Health and Human Services, 2019).

Clinicians in this study did not claim to be seeking or achieving systemic reform through IGMVs, though many envisioned changes in Medicaid and Medicare reimbursement policy (Clemans-Cope et al., 2017; Weeks, 2017) and other policy shifts that would advance health equity, in part through access to CIH. Instead, they found ways to leverage the resources that were available in safety net settings to provide integrative care for chronic pain patients. This resulted in an incremental type of reform, characteristic of tinkering with standards of care, in which “the same problem when raised in different contexts will consequently not give rise to the same solution” (Knorr, 1979)). This in turn resulted in individualized, ad hoc practices that are unevenly applied and therefore limited in the patients they could reach.

At the programs we observed, clinicians were aware of the growing evidence base for CIH but were not always building the specifics of their programs in response to research. Instead, they were hopeful that multimodal care including CIH could benefit their patients and structured their programs in response to bureaucratic uncertainty and available resources. However, the extent to which their efforts can increase access to care remains dependent on state and federal policy; for example, IGMV implementation in the safety-net is likely to be more challenging in states that did not increase Medicaid and continue to limit low-income people’s access to primary care.

Without substantial investment at organizational and policy levels, IGMVs were a form of tinkering, programs that were welcomed by many patients and staff, but were neither well-funded by institutions nor expanded to serve all interested patients. IGMVs create a small crack in the wall of US healthcare, supporting patients with chronic pain through a combination of peer support, extended time with clinicians, access to CIH, and—for some—opioid and other medications. This crack—in the form of workarounds and tinkering—cannot not take the place of systemic reform. The impact of IGMVs and other forms of tinkering is situational and specific.

Clinicians reported that offering care in IGMVs allowed them to practice the kind of medicine they found most helpful, in some cases preventing burnout and early departure from safety-net primary care. Where dissemination of healthcare interventions often begins with the most privileged and trickles slowly, if at all, to those served in the safety-net, IGMVs have the potential to do the opposite: created with and for safety-net patients, the right conditions could grow these multimodal forms of care to serve broad populations in ways that are substantially different from usual health care. .

Fig. 1.

Components of Integrative Group Medical Visits (Illustrator: Wilhemina Peragine).

Highlights:

• Opioid overdose crisis has increased interest in complementary and integrative pain management.

• Integrative group medical visits include complementary and integrative care and peer support.

• Existing health care policy limits access to pain management.

• Integrative group medical visits are a promising chronic pain management model.