|Home | About | Journals | Submit | Contact Us | Français|
Integrating substance use disorder (SUD) services with primary care (PC) can improve access to SUD services for the 20.9 million Americans who need SUD treatment but do not receive it, and help prevent the onset of SUDs among the 68 million Americans who use psychoactive substances in a risky manner. We lay out the reasons for integrating SUD and PC services and then explore the models used and the experiences of providers as they have begun SUD/PC integration in California.
The integration of substance use disorder (SUD) services with primary care (PC) can reduce levels of substance use (Gryczynski et al. 2011; Madras et al. 2009), leading to improved physical and mental health (Madras et al. 2009) and overall healthcare cost savings (Babor et al. 2007; Parthasarathy et al. 2003, 2001). Consequently, the Institute of Medicine (2006), the Office of National Drug Control Policy (2010), and the National Prevention Council (2011) have all called for the integration of services designed to address SUDs into primary care settings, and the 2010 Patient Protection and Affordable Care Act (ACA) will facilitate the integration of SUD services with the rest of healthcare (Buck 2011; Druss & Mauer 2010).
Though it is clear that SUD/PC integration is going to happen, what remains to be seen is how. Reports from the Treatment Research Institute (2011, 2010) have taken important steps toward answering this question by identifying providers throughout the country who have successfully integrated SUD and PC services, and the National Association of Community Health Centers has examined national trends in SUD/PC integration (Lardiere, Jones & Perez 2011). This article will build upon these works by laying out the evidence for SUD/PC integration, identifying trends in SUD/PC integration in California, discussing common barriers to successful integration, and suggesting steps that providers have found helpful in overcoming these obstacles.
The move toward SUD/PC integration fits within the more general trend of increased coordination of services for behavioral health (BH) disorders—particularly low- to mid-level depression and anxiety—with traditional medical care. Decades of research and experience with BH/PC integration have shown that integration can facilitate access to BH services and improve outcomes for certain BH disorders in a cost-effective manner (Collins et al. 2010; Blount et al. 2007).
In the Milbank Memorial Fund’s report Evolving Models of Behavioral Health Integration in Primary Care, Collins and colleagues (2010) identify eight practice models for the integration of BH and PC services that are being utilized across the country:
Though research shows that coordinated and integrated care for depression and anxiety leads to better outcomes, there is no strong evidence concerning the relative effectiveness of particular integration models (Butler et al. 2008).
For many of the same reasons that integration can improve the mental health of clients with mood disorders and anxiety disorders, it also has the potential to benefit individuals with substance use conditions.
Of the 23.5 million Americans who need specialty treatment for SUDs, only 2.6 million—just over 11% of them—actually receive it (SAMHSA 2010a). Stigma is one reason SUD service needs are unmet, as concerns that employers or others in the community have negative opinions of “alcoholics” and “addicts” lead many individuals with SUDs to keep their problems secret (Luoma et al. 2007). Approximately 12% of individuals who recognize that they need SUD services but do not receive them cite concern that others would have a negative opinion if they enter treatment as a reason they do not pursue care (SAMHSA 2010a). But a much more significant cause of the SUD treatment gap is that the vast majority of people with SUDs do not believe they have a problem that warrants professional attention. Almost 95% of those who meet the diagnostic criteria for substance abuse or dependence and do not receive treatment do not think that they need SUD services (SAMHSA 2010a).
This does not mean, however, that these individuals do not seek help of any sort. Many people with SUDs present in medical settings for the treatment of physical or mental ailments that are related—either directly or indirectly—to their substance use behaviors (Ernst, Miller & Rollnick 2007). Over 7.5 million individuals show up in emergency departments for the treatment of problems related to alcohol use each year (McDonald, Wang & Camargo 2004), and approximately 22% of all patients who present in healthcare settings have a substance use condition (Treatment Research Institute 2010). Consequently, medical settings are ideal places to identify individuals with SUDs, engage them in treatment, and begin providing them services (Cherpitel & Ye 2008; Babor et al. 2007).
Medical providers can identify patients who are using substances in a risky manner with brief validated screening tools such as the CAGE and the CAGE-AID, or longer but more comprehensive tools such as the World Health Organization’s ASSIST and AUDIT instruments (Babor et al. 2007). For patients who are at risk, providers can utilize brief intervention techniques, such as motivational interviewing, which elicits behavior change by helping patients explore ambivalence about their substance use (Rollnick & Miller 1995), and the FRAMES model, which combines provider feedback, advice, and empathy to motivate patients to change, give them a sense of responsibility for their substance use behaviors, and build their optimism that they can reduce their substance use (Miller & Sanchez 1994). With these tools primary care providers can address problematic substance use behaviors by helping clients recognize substance misuse and develop the skills and resources they need to change (Madras et al. 2009; Solberg, Maciosek & Edwards 2008; Babor et al. 2007). If systematically implemented as part of a more comprehensive screening, brief intervention, and referral to treatment (SBIRT) program, services provided in primary care settings can be particularly effective in reducing substance use across patient populations. A recent federally-funded initiative to institute SBIRT using SUD treatment providers, BH counselors, community health workers, health educators, and healthcare professionals in a variety of medical settings led to a 67.7% reduction in drug use and a 38.6% reduction in heavy drinking (Madras et al. 2009). Thus SBIRT holds promise as a means to reduce excessive substance use among a significant portion of the patient population (Gryczynski et al. 2011; Kaner et al. 2009; Madras et al. 2009).
SUD/PC integration can also help prevent risky drinking and drug use from developing into more serious problems. There are approximately 68 million Americans who drink alcohol or use drugs in an unhealthy or dangerous manner, but do not meet diagnostic criteria for abuse or dependence (Humphreys & McLellan 2010). Though such individuals do not need specialty SUD treatment, their drinking and drug use can cause significant and permanent changes to the brain’s reward circuitry—alterations that can, in some individuals, lead to SUDs (McLellan et al. 2000). Interventions as brief as one 15-minute session can reduce the frequency and level of substance use (Humphreys & McLellan 2010; Solberg, Maciosek & Edwards 2008; Babor et al. 2007; Kypri 2007), thus helping prevent drinking and drug habits from evolving into more serious disorders.
Reducing levels of alcohol and drug consumption through integrated SUD/PC services can lead to improvements in health domains that are negatively impacted by substance use. Frequent and heavy drinking and drug use often lead to myriad health problems because of psychoactive substances’ toxic effects and the behavioral risks associated with excessive use (Druss & von Esenwein 2006). SUDs are associated with increased risk for hepatitis, sexually transmitted diseases, and gastrointestinal, cardiovascular, pulmonary, renal, hematological, gynecologic, and metabolic problems (Buck 2011; Stein 1999). Furthermore, approximately 20% of adults with a mental illness meet diagnostic criteria for substance dependence or abuse as well (SAMHSA 2010b).
Addressing substance use behaviors that complicate other physical and behavioral problems can help PC providers improve overall clinical outcomes. Integrated SUD/PC services can improve patients’ physical and mental health (Madras et al. 2009), increase the likelihood that patients with HIV will receive antiretroviral therapy (Parry, Blank & Pithey 2007), and decrease hospitalization rates, inpatient days, and emergency department utilization (Parthasarathy et al. 2003; Weisner et al. 2001). Consequently, integrating SUD services with medical care can lead to significant healthcare cost savings for patients with conditions related to substance abuse (Babor et al. 2007; Parthasrathy et al. 2003, 2001). For individuals with medical problems associated with substance use, integrating primary care services provided by physicians, medical assistants, and nurses with SUD treatment can cut total monthly medical costs in half (Parthasarathy et al. 2003; Weisner et al. 2001). Brief physician advice regarding substance use can lead to $4.30 in savings for every dollar spent (Fleming et al. 2002), and in the Washington State Medicaid program, the use of SBIRT in emergency departments has been associated with an average estimated cost reduction of $366 per patient per month (Estee et al. 2010). Gentilello and colleagues (2005) estimate that if trauma centers across the United States were to implement SBIRT protocols, the healthcare system could achieve $1.82 billion in savings each year.
In spite of the potential benefits of SUD/PC integration, it is lagging behind the integration of mental health (MH) services with PC. Whereas over 80% of the nation’s Federally Qualified Health Centers (FQHCs; safety net healthcare providers, including community health centers, public health centers, outpatient health programs, and programs serving migrant and homeless populations) offer MH services on-site, only 55% offer SUD services on-site. (Lardiere, Jones & Perez 2011). While 90% of FQHCs screen for depression, only 65% screen for substance abuse (Lardiere, Jones & Perez 2011). Overall, while MH disorders are just twice as prevalent as SUDs in the U.S. adult population, over six times as many individuals receive MH treatment in FQHCs than receive SUD services (Lardiere, Jones & Perez 2011; SAMHSA 2010a, b).
Though not as widespread as MH/PC integration, efforts to integrate SUD and PC are nonetheless taking shape in California. The work presented here is a first step toward identifying what these initiatives are doing, finding common trends in SUD/PC integration, and laying out the lessons learned, both about the barriers to successful SUD/PC integration and potential ways to overcome them.
To identify SUD/PC integration activities in California, our research team conducted a 39-question survey of county SUD administrators who oversee publicly-funded SUD treatment services in each of the state’s counties.1 The survey, which was distributed in October 2010, was designed to elicit information about SUD/PC integration programs, such as the integration models that are being implemented, staffing for integrated services, and administrators’ thoughts on what barriers need to be overcome to make integration successful. The survey had space provided for respondents to describe their programs in detail, though the majority of data came from multiple choice questions. The multiple choice responses were drafted by the research team based on integration models and barriers observed in the course of evaluation and technical assistance site visits throughout the state over the course of the preceding year. The survey was approved by Institutional Review Boards from the University of California, Los Angeles and the state of California.
We built upon the findings of our survey with site visits and key informant interviews at 11 service provider sites. At each one, we noted models of SUD/PC integration that are being utilized, service delivery procedures, perceived barriers to integration, and solutions that are being implemented to overcome them.
In addition to our site visits, our team partnered with the California Department of Alcohol and Drug Programs (ADP) to convene California’s Forum on Integration, a meeting of SUD administrators and providers along with service integration experts from across the nation, where participants brainstormed to identify successful models and processes for SUD/PC integration, discuss common barriers, and formulate common solutions. Since April 2011, we have also collaborated with ADP to host monthly integration learning collaborative teleconferences, which give providers and administrators continued opportunities to discuss and receive feedback on their integration initiatives.
Throughout California, SUD/PC integration efforts are well underway. Of the 57 SUD administrators who received the integration survey, 44 responded, with 25 (56.8%) reporting that SUD/PC integration efforts are taking place in their county systems. Another 18 (40.9%) reported planning SUD/PC integration within the coming year. To date, counties are primarily using three different models for the integration of SUD services with PC, each of which incorporate aspects of the eight models that Collins and colleagues identified in their description of BH/PC integration practice models:2
Of the 25 California counties integrating SUD services with PC, 23 are doing this in coordination with, or as part of, broader efforts to integrate BH services—encompassing both MH and SUD services—with PC. At most sites using this model, specialty BH providers such as counselors and social workers provide services in PC settings, and are responsible for screening patients and providing brief interventions to address both mental health problems and issues related to substance misuse.
At one site we visited, SUD services are offered under the auspices of the clinic’s integrated BH program, which assigns BH specialists (BHS) to work as members of PC teams. Under this setup, the PC medical doctor maintains responsibility for coordinating and providing most patient care but delivers services in conjunction with the BHS. Patients who receive BH services are initially identified by a self-administered questionnaire to detect possible MH and substance use issues. In the course of one year, this screening tool was able to identify a large number of patients with potential MH or substance use problems, as two thirds of patients screened positive for one or more BH issue, over 44% screened positive for anxiety or depression, and 27% screened positive for potentially problematic use of alcohol or other psychoactive substances.
The questionnaire that each patient at this site fills out is then reviewed by the treating physician. If, upon talking to the patient, the physician thinks that the patient needs further assessment for a MH disorder or SUDs, the BHS is called in. The BHS conducts screenings and interventions in the same exam room that is used for the delivery of other medical services, thus making it apparent to the patient that the behavioral health staff is an ordinary member of the medical team. If the BHS’ assessment reveals that the patient needs more services, the BHS then provides interventions onsite, including supportive counseling, psycho-education, motivational enhancement, behavior change skills training, and help with self-management. Generally, behavioral interventions last up to three sessions, and if more intense services are needed, patients are referred to either an in-house medical social worker or offsite for specialty behavioral health services.
At many clinics that utilize this model, however, the procedures for identifying and treating SUDs are not as systematic or precise. In several locations using BH professionals to provide SUD services, screening for substance use is relatively informal, not done universally, and not focused on special target populations (such as pregnant women or individuals with liver disease). Furthermore, screening protocols at many of these sites often do not involve the use of validated SUD screening instruments, and are only conducted when a PC provider has a “hunch” or “suspects” that patients may have an alcohol or other drug problem. Consequently, many PC providers—especially those who are not trained on how to identify patients who drink or use drugs in a dangerous manner—may fail to identify many patients who need SUD services or are at risk because of their substance use behaviors. Furthermore, since the orientation of most of the BH staff at these sites is more geared toward MH than SUD treatment, the services they provide tend to focus more on managing psychological health than on addressing substance use behaviors. Providers at these sites requested further training so they could improve their screening protocols and identify more patients in need of brief interventions or formal treatment to address their substance use.
Eleven (44%) of the 25 California counties currently integrating SUD and PC services are using models of reverse co-location, placing PC providers in specialty SUD treatment settings. Some of the most common settings for this approach are licensed methadone clinics (or narcotic treatment providers [NTPs]). Unlike many specialty SUD programs, which traditionally have little medical staffing, NTPs are required to have medical personnel onsite to determine medical eligibility for methadone and ensure that it is administered in a safe and effective manner. Since medical staff are already established in NTPs, the integration of PC services designed to address patients’ other medical needs has been a relatively smooth process. Co-located PC providers in NTP settings offer a variety of services, ranging from routine physicals and screening for chronic diseases to lifestyle counseling and referrals to medical specialists when necessary. SUD providers using this model report that offering medical services on-site is particularly helpful for NTP clients, many of whom are low-income and have little access to medical care elsewhere.
At one SUD site we visited, medical care was being integrated into the array of services offered to residents in an inpatient SUD treatment program. Traditionally, it has been difficult to bring PC services into inpatient SUD settings since licensed residential facilities cannot employ staff in any formal medical position. However, by using nonmedical staff to screen for physical health problems and link patients with appropriate medical services, the facility has come up with a way to identify and serve the medical needs of its residential clients. An on-site staff member at this facility coordinates residents’ medical services with outside providers, facilitating referrals in order to ensure that patients’ physical health is properly addressed.
Ten (40%) of the counties that have begun integrating SUD and PC services are using models that place SUD treatment specialists within PC settings. In nine of the ten counties, SUD staff conduct screenings for risky alcohol and drug use, and in seven counties, they also provide brief intervention and psychosocial SUD treatment services on-site. In most counties using this model, SUD staff are partially integrated into the medical clinic, collaborating with PC staff, but still using separate systems for billing and documentation. Only three counties reported that SUD treatment staff were fully integrated—clinically, administratively, and financially—with the rest of the primary care practice.
In the course of our site visits, we observed that the placement of SUD staff into PC clinics has led to the identification and treatment of many patients whose SUD service needs were previously undetected. At one FQHC we visited, all individuals who receive care at the site—be it for physical health, mental health, sexually transmitted diseases, or domestic violence—are systematically screened for substance use. Those who screen positive are then physically linked to the on-site SUD specialist through a “warm handoff,” with clinic staff walking the patient down the hall and personally introducing them to the SUD staff member. By doing this, the clinic is able to both reduce the discomfort that patients may feel when told they need to see an SUD specialist, and ensure that they will follow through with their referral for further assessment and services. With this system, the clinic reports that over 90% of patients who screen positive on their initial SUD screening tool are successfully linked to the on-site SUD specialist. Upon meeting with a new patient, the SUD specialist conducts further assessment to determine the need for treatment. If further services are needed, the SUD specialist provides them on-site, and the SUD specialist also consults regularly with PC staff on matters regarding patient care at weekly case conferences.
In spite of its benefits, this model may not lead to successful screening and service provision if there are communication problems between SUD staff and other providers. At one FQHC we visited, the county SUD department placed a certified counselor in the FQHC to conduct screenings, provide brief interventions, and give referrals to specialty SUD services when necessary. However, out of approximately 9,000 patients who received services at the FQHC over the course of a year, less than 100 were assessed for substance use or abuse problems, and attendance at on-site SUD groups was extremely low. By contrast, approximately 1,500 patients at the clinic received mental health services during the same time period. The cause of the problem, we learned, was not the integration model itself, but rather its implementation. Most of the clinic’s staff were unaware of the SUD specialist’s role in the FQHC, and since the SUD specialist’s office was at the end of a long corridor that separated it from the rest of the clinic, neither patients nor providers could interact with the co-located SUD staff easily. Furthermore, clinic doctors reported that it was difficult for them to discuss substance use with their patients, and that when they tried to make referrals to specialty SUD services, they never received updates on the patient’s SUD diagnosis and progress.
Once they became aware of these problems, both the FQHC medical staff and the co-located SUD specialist were able to rectify them quickly. Clinic staff were informed of the availability of on-site specialty SUD services and educated in motivational interviewing and other techniques to engage reluctant patients in SUD treatment. The co-located SUD specialist began completing disposition forms that were returned to physicians to give them updates on patient progress, and also began providing small incentives to encourage client attendance at on-site groups. Within months of these small changes taking effect, screening and service utilization increased dramatically. Rates of SUD assessment doubled in less than a year, and attendance at on-site SUD groups grew from almost zero to near capacity within two months. Once the communication problems at this FQHC were cleared up, some of the benefits of SUD/PC integration—better identification of substance use problems, and better access to SUD services—were realized. With these successes, however, other challenges have emerged. As the clinic has identified increasing numbers of clients in need of SUD services, demand has overwhelmed capacity, and providers have had difficulty accommodating all clients who need to attend groups and/or receive detoxification treatment. Thus, as experience at this site shows, even when the initial challenges in implementing integration protocols are overcome, scarcity of SUD treatment resources can limit the reach of integrated services.
According to SUD administrators in California, the most common barriers to SUD/PC integration involve financing, documentation, and partnering with PC providers.
The most commonly reported barrier to SUD/PC integration in California is inadequate or inflexible funding. Nearly all (92%) of the counties that have begun integrating SUD and PC services report that finding ways to adequately and sustainably finance integrated care is a major concern. To a certain degree, regulations at the state level are to blame for this. California’s Medicaid program does not reimburse for substance use screening and brief intervention services, nor does it allow FQHCs to bill for both physical health and behavioral health visits provided to an individual patient on the same day. Though these are apparently administrative technicalities at first glance, they are major obstacles to making SUD/PC integration feasible for patients enrolled in the state’s Medicaid program, as PC providers struggle to absorb the costs of uncompensated SUD screening and brief intervention services. Administrators also cite lack of reimbursement for collaborative care and case management related to SUDs as well as for services provided by unlicensed clinicians who could otherwise provide SUD services as barriers to integration. Though policy changes under the ACA and California’s 1115 Medicaid Waiver may help overcome some of these obstacles in the future, they nonetheless represent major roadblocks for administrators and providers who are trying to integrate SUD and PC services today.
Another set of major barriers to SUD/PC integration reported by SUD administrators involves documentation. In the integration survey, 63% of respondents indicate that documentation issues hinder integration activities. In particular, Federal Regulation 42 C.F.R. Part 2 is cited as a serious impediment. Under this regulation, records regarding SUD treatment in designated SUD programs cannot be shared with other health providers or included in open electronic health records without written consent. Many administrators and providers remain unclear about whether or not SBIRT or treatment that takes place in PC settings is subject to the same restrictions. Moreover, administrators are concerned that if SUD and PC treatment providers cannot freely and openly share information regarding SUD treatment, the potential clinical benefits of integration could remain unrealized. Though many of these problems could be solved by having patients sign consent forms authorizing the sharing of SUD records before treatment, or by modifying electronic health record software, many providers are still unsure of how to make these changes while still satisfying federal privacy requirements.
A third set of barriers that have emerged in the course of SUD/PC integration in California concerns the challenge of forming effective partnerships with PC providers. In the integration survey, 63% of respondents reported this concern, and much of the discussion at ADP’s integration forum focused on these issues. PC providers, who are already balancing large caseloads and packed schedules, are often reluctant to implement new protocols (such as SUD screening) since they have limited face-to-face time with patients and prefer to focus on more immediate medical concerns. Administrators also report that some PC providers stigmatize individuals with SUDs, harbor misconceptions about SUD medications, and are suspicious about working with specialists from other disciplines.
In spite of these challenges, many providers in California have been able to integrate SUD and PC services. At the integration forum, stakeholders who have overseen successful SUD/PC integration initiatives shared their insights on how to overcome some of the most common obstacles that stand in the way of effective SUD/PC integration.
To overcome financing barriers, participants at the integration forum agreed that one solution is to initiate SUD/PC integration projects with smaller, but more accommodating, funding sources. Grants from governments, foundations, and nonprofit organizations can fund new integration programs free of concerns about federal reimbursement. Flexible forms of state and local funding can also help overcome some of the requirements and regulations that inhibit SUD/PC integration. Prevention and Early Intervention funding from the California Mental Health Services Act, for example, has been used to start up integration initiatives in several counties. Though often not substantial enough to facilitate large-scale or systemwide integration, funding that is not reliant on Medicaid has helped administrators establish the infrastructure, protocols, and procedures for SUD/PC integration at the local level.
Forum participants also identified many concrete steps to facilitate partnerships for SUD/PC integration, both across systems and within service delivery organizations. At the county level, presenting research and data illustrating the potential benefits of integration has enabled SUD administrators to increase buy-in from potentially skeptical PC partners. Administrators have also been able to foster successful partnerships through activities that require PC and SUD organizations and providers to work in collaboration. Joint trainings on integrated care, for example, give PC and SUD providers opportunities to get to know each other and work together. In one county, an administrator reported that by requiring BH and PC organizations to collaborate when applying for public funding to support integration projects, the county has been able to facilitate and foster mutual trust and strong working relationships between providers from different systems.
At the clinic level, administrators and providers have also identified several steps that can facilitate successful integration. When integrating SUD services into PC settings in particular, forum participants highlighted the importance of finding “champions” for integration—well-respected providers who value, and are willing to advocate for, changes to organizational structure and clinical protocols. When models involving co-location are used, forum participants agreed that it is critical to establish effective channels of communication between PC and SUD providers, so that differences in clinical and administrative cultures and practices can be easily identified and overcome. To minimize resistance to integration and make it more palatable to clinical staff, forum participants recommended implementing integration initiatives gradually, starting with simple co-location models and then, over time, instituting measures to enhance administrative and clinical integration. Another strategy that has been successful in several counties has involved taking steps to avoid adding too much to the workload of clinical staff and physicians. For example, using front desk staff to perform initial SUD screenings, or simply adding SUD screening questions to pre-existing health and mental health screening protocols, has helped many clinics ensure that the addition of integrated SUD services does not pose too heavy a burden on providers.
While the lessons learned about SUD/PC integration thus far are good starting points, there are still as many questions about SUD/PC integration as there are answers. Research is needed to identify what models for SUD/PC integration are most effective, and which are most appropriate for specific patient populations and treatment settings, in order to inform both policy and clinical practice. While further experience and research will help identify best practices and models for SUD/PC integration, some of the early lessons learned about SUD/PC integration—its potential value, potential barriers to it, and potential solutions to overcome those barriers—will, it is hoped, guide providers as they begin to transform the vision of SUD/PC integration into a reality.
1Two rural California counties provide SUD services under one shared system, so there are only 57 county SUD administrators in the state even though there are 58 counties.
2Other promising and highly innovative SUD/PC integration models—such as San Francisco’s Office-Based Opiate Treatment program (Hersh, Little & Gleghorn 2011)—are being implemented in California as well. However, the three models discussed here are the most common ones being used throughout the state.