Applying lessons from task sharing in global mental health to the opioid crisis

Abstract

The current opioid crisis in the United States has been considered an “epidemic of poor access to care”. Similar to the shortage of trained providers to prescribe medications to treat opioid use disorder (OUD), there is a severe shortage of trained providers to meet the mental health needs of patients with OUD. These workforce shortages are evident nationwide, yet are particularly salient in rural areas. In this commentary in response to the article “Integrating Addiction Medicine into Rural Primary Care: Strategies and Initial Outcomes (Logan et al., 2019),” we propose that we can apply lessons learned from working in resource-constrained settings globally to improve access to mental health care for patients with OUD in rural areas in the US. We expand upon Logan et al. 2019 by discussing how non-specialist health workers, including community health workers and peer providers, under the supervision of psychologists and other specialists, can expand access to evidence-based mental health care for patients with OUD, particularly those receiving medications for opioid use disorder (MOUD). We draw from established models in global mental health that rely on “task sharing” mental health care to discuss ways in which lessons learned from scaling up evidence-based interventions with lay health workers in low and middle-income countries can directly inform efforts to increase access to mental health care to address the opioid crisis.

Nationwide the opioid crisis has been considered an “epidemic of poor access to care” (Wakeman & Barnett, 2018). This “epidemic of access” typically refers to a shortage of prescribers of medication to treat opioid use disorder (OUD). Yet, also in short supply are providers of evidence-based psychosocial interventions to treat the psychiatric comorbidities that frequently co-occur with OUD. Approximately 40% of individuals in the US with any type of substance use disorder (SUD) have a comorbid psychiatric disorder, yet fewer than half (48%) receive treatment for either disorder (U.S. Department of Health and Human Services, 2016). The uneven distribution of the behavioral health workforce exacerbates this gap; most professionals are concentrated in urban areas and the private sector, leaving much of the population, specifically rural areas and often the poorest and most vulnerable people, without access to care.

Logan et al., in press, in their article “Integrating Addiction Medicine into Rural Primary Care: Strategies and Initial Outcomes” conducted a retrospective review of outcomes of an integrated buprenorphine and behavioral health program based in primary care in a rural, underserved community. They reported on retention, substance use, and mental health outcomes (depression and anxiety) over three months among patients who were prescribed buprenorphine in the integrated primary care program. Although there is growing data that medication for opioid use disorder (MOUD) alone without concurrent behavioral therapy is an effective treatment for OUD (D’Onofrio et al., 2015), the findings of Logan et al., in press demonstrate high rates of mental health comorbidity among patients receiving buprenorphine for an OUD, with 77% having a moderate to severe comorbid mental health condition. Their findings also highlight that most patients also have at least one other substance use disorder diagnosis (predominantly alcohol and methamphetamine in their population) in addition to OUD. Similar to previous research (Carroll et al., 2018; Eastwood, Strang, & Marsden, 2019), this study found that mental health and other substance use comorbidity was associated with premature dropout from the MOUD program. This finding suggests that for MOUD to be effective, programs should also provide patients with treatment for their mental health and other substance use comorbidities. These efforts will demand greater integration of evidence-based behavioral interventions into MOUD programs to treat both mental health (Carroll et al., 2018) and other SUDs for which there is not effective medication treatment available (e.g., for stimulant use disorders, such as methamphetamine use disorder; Lee & Rawson, 2008). Overall, we suggest that the findings of Logan et al. in press, make a compelling case for the importance of behavioral interventions alongside MOUD for three main priorities: (1) to support retention in care; (2) to address mental health comorbidities; and (3) to reduce the severity and harm of other SUD comorbidities.

Yet, taking behavioral interventions to scale can be challenging, particularly in rural areas where behavioral health professionals are often scarce. We suggest that low-cost, brief behavioral interventions, such as cognitive behavioral therapy (CBT) and behavioral activation (BA)-based interventions, may allow MOUD programs to treat the intersection of mental health and substance use comorbidities, and improve retention. Importantly, these interventions have been feasibly delivered by lay health workers (LHWs), including peers and community health workers (Magidson et al., 2017). These approaches, delivered by LHWs, have the potential to address mental health provider shortages (e.g., psychologists and other specialized mental health providers) in underserved and rural areas most affected by the domestic opioid crisis.

We propose that we can draw from experiences training LHWs in resource-limited settings globally to deliver these behavioral interventions to patients with OUD. For instance, our team is leading a series of studies conducted in resource-limited clinics in sub-Saharan Africa to integrate mental health and substance use interventions into HIV and other chronic disease care ( PI: Magidson; Myers et al., 2018). In these ongoing trials, we are using “task sharing” approaches to train LHWs in evidence-based CBT interventions. Task sharing models involve training less specialized health care workers in mental health care delivery, with a focus on distributing and delineating tasks across a larger team, including specialists who provide ongoing supervision, support, and consultation to LHWs (Hoeft, Fortney, Patel, & Unützer, 2018; Magidson et al., 2017). From implementing these trials of LHW-delivered interventions using task sharing models, we have learned many key lessons that can be directly applied to expanding behavioral health services in the domestic opioid crisis.

First, we have learned that “task sharing” does not negate the role of the specialist; rather, the specialist needs to learn to become a supervisor to the other providers and to create referral pathways so that people who need higher levels of care can access it (Hoeft, Fortney, Patel, & Unützer, 2018). In our trials, the intensive training, supervision and fidelity monitoring for LHWs (typically by a clinical psychologist or registered psychological counselor) is essential to ensure adherence to the intervention, quality care, and patient safety. Similarly, clinical psychologists in the US will play an important role in training and supervising LHWs, whose roles are expanding to meet the behavioral health needs of patients with SUD (Bassuk, Hanson, Greene, Richard, & Laudet, 2016). Specifically, we propose that clinical psychologists have an important role in (1) developing behavioral interventions that can be feasibly delivered by LHWs to expand access to treatment for patients with OUD; (2) developing models that provide LHWs with adequate support and supervision, including personal debriefing; and (3) addressing complex, treatment resistant patients in stepped care models and/or supporting LHWs to address clinical emergencies to ensure patient safety.

Specifically, peer recovery coaches—or individuals in recovery from SUD who are hired as part of the care team to support patients with SUD—are being scaled as part of efforts to increase access to SUD treatment in the US (Bassuk et al., 2016). As peer recovery coaches assume a greater role in the response to the opioid crisis to meet the behavioral health needs of individuals with OUD, we need to create referral pathways, supervision and support structures, while building the capacity of specialists to respect and support this new type of worker. Indeed, there is a lack of clarity on how peer coaches for OUD should be trained and supervised, particularly with regard to the more behavioral aspects of their job. In a qualitative study our team conducted on peer recovery coaches for SUD in the US, we found that coaches did not feel adequately prepared to manage the prevalent mental health symptoms of their patients; they also did not always feel supported by the medical providers, worrying that their lived experience was not taken as seriously as medical degrees (Jack, Oller, Kelly, Magidson, & Wakeman, 2017).

Second, we know from our ongoing trials in South Africa ( PI: Magidson; Myers et al., 2018) and other global mental health trials (Singla et al., 2017) that LHWs can feasibly and effectively deliver more structured, evidence-based interventions, including behavioral activation (Magidson et al., 2015; Nadkarni et al., 2017; Patel et al., 2017), problem solving, and motivational enhancement therapies (Myers et al., 2019; Myers, Stein, Mtukushe, & Sorsdahl, 2012), with regular supervision, training, consultation, and monitoring to ensure fidelity to the intervention model (Magidson et al., 2017; Murray et al., 2011). A recent meta-analysis evaluating LHW-delivered CBT interventions in low- and middle-income countries (LMICs) demonstrated moderate to strong effects on common mental disorders, including depression and anxiety disorders, with a pooled effect size of 0.49 (95% CI: 0.36–0.62) (Singla et al., 2017). Although for peer recovery coaches in the US speaking with patients about behavior change and drawing from their own experience is a large part of their role, individual counseling is typically less structured, and formal mental health counseling is often not within their scope of practice or job description. We propose that task sharing models from LMICs may provide an empirical basis for expanding peers’ current role in the US to include delivery of structured, evidence-based interventions, such as those we are evaluating in LMICs (i.e., behavioral activation, problem solving). This has important policy implications for peers’ scope of practice as certification, training, and funding legislation is being prioritized nationwide. To advance the field, researchers and clinicians conducting this work should describe in greater detail, and subsequently evaluate, effective models for training and supervision (Murray et al., 2011). It is essential for supervisors to consider how to balance training and supervision to promote adherence to more structured approaches, while ensuring that peers continue to feel agency over the care they provide and can incorporate their lived experience appropriately.

Indeed, while expertise through lived experience is powerful, it may be enhanced with structure, supervision, and a foundation both in theory and empirics to be replicable and effective (Bassuk et al., 2016). Despite the rapid, recent scale-up of peer recovery coach programs in the US, the empirical support for these programs has lagged behind their expansion, and a recent review points to the need to improve the specificity of peer-delivered interventions and study methodology (Bassuk et al., 2016). Members of our team (JM, HJ) have evaluated the effect of peer recovery coaches integrated into primary care on health care utilization and buprenorphine treatment outcomes (Magidson et al., 2018). Our results demonstrated a significant decrease in costly acute care utilization and a significant increase in engagement in outpatient services, including mental health treatment, following recovery coach contact over a one year period. In a subset of patients prescribed buprenorphine, current recovery coach contact more than doubled the odds of buprenorphine treatment retention and increased the odds of opioid abstinence (Magidson et al., 2018). Our team’s work suggests that recovery coaches may be an impactful and potentially cost-effective addition to the SUD care team; however, additional research using a matched comparison condition is needed, and research evaluating peer recovery coach delivery of evidence-based psychological interventions.

Peer coaching programs for OUD should work to balance the sharing of coaches’ personal experiences with additional evidence-based strategies. This tension is evident in Logan et al.’s study in that there were distinct roles within the behavioral health team; behavioral health providers (psychologists and a licensed clinical social worker) provided the structured, evidence-based counseling, and a separate behavioral health case manager (BHCM) supported other wraparound service needs, including housing and accessing community resources. In settings that lack access to specialized mental health providers, or those that cannot fund or recruit behavioral health providers to fill separate and distinct counseling and case management services, we may consider a task sharing model with peers or other LHWs to offer an integrated solution for both. Efforts to integrate structured counseling and sharing of lived experiences among peer providers will require attention to understanding key ingredients necessary for effective peer-delivered behavioral interventions to support MOUD outcomes.

Finally, another key lesson from our work in South Africa has indicated a patient preference for peer providers. In our qualitative work in South Africa (Magidson et al., 2019), when we asked patients who they preferred to meet with to support substance use-related behavior change, they indicated a preference for working with peer providers. As we also have found in the US, by bringing their own lived experiences to their interactions with patients, peer providers can create a space that patients perceive as less stigmatizing and more comfortable than interactions with other health care providers (Jack, Oller, et al., 2017). Having providers who are in recovery themselves may further amplify the benefits of integrating psychosocial interventions into MOUD programs, as peers’ lived experiences may strengthen their therapeutic connection with patients and further destigmatize OUD in clinical and community settings. Logan et al., in press found that the “visibility and consistency” of the BHCMs appeared to help “establish trust” and that they had a “unique opportunity to influence institutional perspectives and increase access to care” through “destigmatizing and treating SUDs.” Based on our prior work, we believe that embedding a peer provider may further amplify the destigmatizing effects of integrating OUD care programs. Peer coaches in the US can guide professional providers on the needs and cultural practices of a community, potentially helping combat the stigma of opioid use within the healthcare system (Bassuk et al., 2016). This peer role may be particularly important in a setting such as the one that Logan et al., in press describe. In this rural, western part of Hawai’i Island, where a majority of the patients are Native Hawaiian, Pacific Islander and low-income, there is the potential for significant demographic differences between providers and patients that peer providers could help bridge.

Peers also may help expand the “reach” of MOUD to support patients with OUD. For instance, our team (JM)’s community-based work in Baltimore focuses on using a peer recovery coach to support individuals with OUD who are not engaged in health care services and may be hesitant or face barriers to initiate MOUD (Satinsky et al., 2019). In this work, the peer recovery coach balances sharing of his own personal story to increase motivation and reduce stigma, while also utilizing evidence-based strategies such as motivational interviewing, behavioral activation, and problem solving to encourage the initiation of OUD services and retention in care. Similarly, bringing care into communities and a greater presence of peers in recovery could help the areas hit hardest by the crisis begin to heal and to fight community-level stigma.

In underserved settings hardest hit by the opioid crisis, there is an urgent need to expand access to treatment for OUD. Given that MOUD as a standalone treatment can reduce opioid use (D’Onofrio et al., 2015), there is great need to increase access to MOUD regardless of whether other behavioral health services are available. However, Logan et al., in press showed that engagement for MOUD may be improved with behavioral interventions, as these interventions can increase engagement in care. Further, and perhaps more importantly, the vast majority (over 75%) of patients on MOUD in their study had co-morbid mental health conditions and other SUDs, and providing behavioral interventions for these conditions alongside MOUD is likely to enhance OUD outcomes as well as improve patient quality of life.

One potential barrier to treating OUD in rural areas is the lack of highly trained providers or specialists to consult with about challenging cases and to provide clinical supervision and mentoring to less experienced providers. In rural areas most impacted by the OUD crisis, there may be an important role of telehealth in improving access to care and building provider capacity (Chakrabarti, 2015; Hilty et al., 2013; Mehrotra et al., 2017). Online platforms have emerged as a useful strategy for increasing primary care providers’ competence in treating SUDs and may also provide a platform for extending the reach of behavioral interventions alongside MOUD (Komaromy et al., 2016; Sockalingam et al., 2018); Project ECHO, a national, telehealth, hub-and-spoke model, facilitates both didactic trainings and open-ended case discussions for rural providers treating SUDs, led by more experienced or specialized physicians. Project ECHO sessions are online, in real time, and often conducted during a lunch break in a clinical day (Komaromy et al., 2016). While Project ECHO has been aimed at medical providers, a similar model could also be used for psychologists to support other behavioral health providers treating OUD.

Policy implications

As the opioid crisis puts pressure on the US to expand access to care while cutting costs, practitioners and policymakers should look broadly for solutions. We should study and adapt evidence-based, low-cost, and efficient approaches that have been developed in resource-constrained settings globally and delivered by LHWs. These approaches may allow for cost-effective delivery of interventions that can help address both the medical and psychological aspects of caring for people with OUD. As demonstrated in Logan et al., in press, these services should be integrated into primary health services where they can most easily be taken to scale within existing infrastructure. Policymakers should promote the peer workforce and change funding for these roles to make it easier to develop these programs and sustain them through billable peer services, which has long presented challenges for domestic LHW programs (Jack, Arabadjis, Sun, Sullivan, & Phillips, 2017). Funding a LHW program within a fee for service model of healthcare can be difficult, and these programs have typically been funded through short-term grants. However, with the growth of global payment models, opportunities may open to fund innovative programs such as these. Now is the moment for funders and policymakers to create flexibility and capitol for clinics, such as the West Hawaii Community Health Center in Logan et al., in press, to support innovative programs to fight the opioid crisis.

Conclusions

The current opioid crisis in the US has disproportionately affected low-income, rural communities. In these settings, it is particularly difficult to recruit SUD treatment professionals, further limiting access to OUD care in regions where geographic access barriers are prevalent. This treatment gap stems primarily from a shortage of health care resources, but is also widened by stigma towards mental health and SUDs, and the poverty that too often accompanies these disorders. The human resource gap highlights how efforts to integrate buprenorphine into primary care, such as those in Logan et al.’s study, are crucial, particularly in rural communities that face the greatest workforce shortages and OUD burden. Further, this study demonstrates the important role that mental health comorbidity has on retention in MOUD programs, suggesting that feasible, scalable, and appropriate behavioral interventions to alleviate mental health symptoms could enhance the impact of MOUD. LHW models in global health and global mental health may provide an important guide for how to feasibly and sustainably expand access to evidence-based services in the opioid crisis.

Public Health Significance Statement.

The opioid crisis is considered an epidemic of poor access to care, which includes poor access to medication for opioid use disorder and treatment for mental health comorbidities. Lessons learned from lay health worker interventions in low-resource international settings could help expand access to OUD treatment locally.