Development of NAMASTE (New Anxiety Management Algorithm Standardizing Treatment Experience) and Implementation in Primary Care

The availability of public health data from surveillance of anxiety disorders in general, and in high-risk populations in particular, falls far short of the availability of surveillance data from monitoring depression (1). Anxiety disorders are as common and possibly more common than depression among patients seen in primary care (2). Comorbid depression and anxiety are associated with higher rates of common medical conditions, higher medical utilization, slower recovery, and greater disability. Furthermore, up to 20% of primary care visits occur during an episode of anxiety or depression, and over 50% involve a comorbid depressive or anxiety disorder. Anxiety disorders pose an independent risk for suicide, and the Centers for Disease Control and Prevention recently reported an approximately 30% rise in suicide over the past decade (3). Benzodiazepines, a medication class commonly used in anxiety disorder management, were detected in approximately 30% of postmortem autopsies of suicides between 1999 and 2015 (4). Next to opioids, benzodiazepines are the most common medication class involved in unintentional overdose.

Prior studies, including IMPACT (Improving Mood—Promoting Access to Collaborative Care Treatment) (5), an evidence-based integrated depression treatment for primary care, have revealed the possibility of achieving positive outcomes in depression care by integrating behavioral health into primary care. Often, such integration is necessary due to the prevalence of depression and the shortage of external care partners for diagnosis and treatment. The University of North Carolina’s Internal Medicine Clinic (IMC) developed and implemented protocols consistent with several randomized controlled trials that have demonstrated the effectiveness of collaborative care for treatment of anxiety disorders through both office-based and electronic application of collaborative care components (6–9).

NAMASTE Design

A better understanding is needed of how to extend effective trials of the components of collaborative care for anxiety disorders into naturalistic and pragmatic implementation. IMC’s NAMASTE (new anxiety management algorithm standardizing treatment experience), a model to integrate anxiety disorder identification and treatment into primary care, and its anxiety treatment program deepen understanding of how to apply collaborative care components—stepped care and integration of behavioral health professionals—to the primary care team in order to co-manage anxiety disorders. Unique characteristics include implementation and evaluation of collaborative care for anxiety disorders through attainable and feasible quality improvement methods, development of a medical education training module, reporting of primary care provider treatment confidence, and reporting of uptake of strategies to address benzodiazepine overuse.

The IMC’s anxiety care program was established to address an unmet need in an academic primary care setting serving over 12,000 patients across North Carolina. The outpatient IMC encompasses over 100 providers, including physicians, clinical pharmacists, nurses, physicians’ assistants, licensed clinical social workers, resident trainees, and multidisciplinary students. The clinic is recognized as a level 3 patient-centered medical home and has been a leader in conducting formal quality improvement activities (10).

In 2010, the IMC implemented an enhanced care depression care program to incorporate algorithmic screening and treatment on the basis of the IMPACT model (5). Applying findings from the substantial research on screening, evidence-based treatment, and program design for depression in primary care medical settings, IMC staff designed NAMASTE to include extended onsite behavioral counseling with psychiatric consultation for patients with challenging presentations. The IMC had previously incorporated U.S. Preventive Services Task Force recommendations to screen for and treat depression, including by addressing screening concerns for suicidal ideation and ensuring adequate availability of onsite integrated behavioral health support.

By 2012, the IMC’s successful treatment of depression raised awareness of previously unrecognized and untreated anxiety disorders. In 2013, the IMC depression care team added a consulting psychiatrist and used interdisciplinary collaboration to develop algorithms for integrated anxiety treatment and to design training informed by existing literature and clinical experience with best practices in a successful depression care program. NAMASTE, formally introduced in 2015, built on the structure of IMPACT and the work of the personnel who had integrated IMPACT-informed depression treatment at the IMC.

The IMC’s depression care was largely guided by use of the nine-item Patient Health Questionnaire (PHQ-9). The anxiety treatment program used the seven-item Generalized Anxiety Disorder Scale (GAD-7) to assess anxiety and severity and to guide measurement-based care. Three intervention products were developed: a treatment algorithm, a medication chart, and case-based learning module and materials. (Initial and follow-up treatment protocols, medication chart, and the case-based educational module are available on request from the authors.) The GAD-7 differs from the PHQ-9 in that it screens for several forms of anxiety disorders. Using a threshold of 10 points, the GAD-7 has a sensitivity of 89% and a specificity of 82% for generalized anxiety disorder. It shows moderate accuracy in screening for three other common anxiety disorders: panic disorder (sensitivity 74%, specificity 81%), social anxiety disorder (sensitivity 72%, specificity 80%), and posttraumatic stress disorder (PTSD; sensitivity 66%, specificity 81%). The GAD-7’s sensitivity to change was generally good (2), and the scale showed early benefit for addressing anxiety in a complex primary care population.

NAMASTE’s algorithm was vetted in consultation with physicians in clinic management as well as psychiatry and primary care providers. The treatment algorithm included a companion medication chart to guide first- and second-line pharmacotherapy, help physicians avoid activating selective serotonin reuptake inhibitors (e.g., sertraline) and serotonin-norepinephrine reuptake inhibitors (e.g., venlafaxine), and provide options for “as-needed” medications. The chart offered guidance for safer as-needed medication options, specifically alternatives to benzodiazepines to manage acute exacerbations of anxiety disorder symptoms and recommendations to safely manage benzodiazepine optimization, tapering, and discontinuation. As part of NAMASTE, clinicians used the GAD-7 when patients were previously diagnosed with major depression or with any kind of anxiety disorder or when physicians suspected such an anxiety diagnosis on the basis of their interview and exam. Providers were encouraged to repeat measurements of the GAD-7 and the PHQ-9 in follow-up visits for any patients with anxiety disorder.

The IMC developed a new anxiety treatment training module to introduce NAMASTE with case-based didactic materials, to present a constellation of adverse health impacts of untreated anxiety disorders, and to demonstrate how to identify and care for affected patients. The training module included educational content and clinic-specific information to help clinicians integrate the treatment algorithms, medication chart, onsite behavioral counseling, and psychiatric assessment and consultation options into anxiety treatment. Annual training was provided in a faculty meeting for attending physicians and in a preclinic conference series for residents. The training session provided evidence of the GAD-7’s ability to identify generalized anxiety disorder, panic disorder, PTSD, and social anxiety disorder. The consulting psychiatrist provided a didactic lecture for attending physicians on identification of anxiety disorders and on treatment and management features of PTSD, obsessive-compulsive disorder, social anxiety disorder, and generalized anxiety disorder. The algorithm and training module highlighted psychotherapy recommendations for PTSD in particular. On the basis of evidence in the literature, the training module encouraged early use of medication along with counseling for patients experiencing anxiety disorders. The consulting psychiatrist conducted a didactic session for attending physicians on prescribing trends, safety data, and indications approved by the U.S. Food and Drug Administration for commonly used benzodiazepines and explained preexisting benzodiazepine equivalency and taper and crosstaper strategies. Treatment for acute periods of heightened anxiety symptoms through pharmacological management was addressed with a pragmatic medication chart to indicate the dosage range and indications for nonbenzodiazepine options. The training module also included this information. Further quality improvement projects related to the anxiety treatment program examined the clinic’s success rate at tapering benzodiazepines and transitioning patients to other medications, such as buspirone, gabapentin, and hydroxyzine. Given the challenges of prescribing controlled substances, colleagues and residents showed strong comprehension of the use of benzodiazepine alternatives.

GAD-7 forms were stocked with other forms in exam rooms and were printed on purple paper for visibility and to aid tracking until responses were entered into the electronic medical record. GAD-7 scores were categorized as ≤9, no or mild anxiety (retest with the GAD-7 as needed); 10–14, moderate anxiety (with direction to treat and retest); and ≥15, severe anxiety (with direction to treat and retest). Follow-up visits were to be scheduled within 4 weeks for patients with severe anxiety and within 12 weeks for patients with moderate anxiety consistent with depression treatment return intervals. The algorithm guided clinicians through consideration of contributors to the disorder, expanded differential diagnosis, options for treatment approach, and guidelines for when to initiate medication and counseling.

The IMC screens for anxiety when patients are identified as having depression—about 10%−15% of the IMC’s patients (N=1,200)—and when an anxiety disorder is suspected. Anxiety is estimated to be comorbid with depression for about 50% of patients and present in 10%−20% of primary care visits. Although the burden of disease on patients is clinically significant, and sometimes life-threatening, the absolute number of patients requiring treatment is manageable.

For comprehensive integrated behavioral health treatment, the IMC offers onsite behavioral counseling and psychiatric consultation and assessment. Counseling helps patients develop anxiety management skills and uses cognitive-behavioral therapy (CBT), problem-solving treatment, and mind-body medicine skills training to provide psychoeducation about anxiety and the relevance of self-care. Psychiatric consultation offers physicians direct counsel with a clinician and single-patient diagnostic consultation with one to three follow-up appointments. Primary physicians maintain responsibility for medication management. In-clinic services help refer patients to outside continuity counseling and psychiatry as necessary.

Quality Improvement

The IMC used the model for improvement developed by the Institute for Healthcare Improvement and small tests of change to refine NAMASTE’s processes and content. The clinic also used the continuous quality improvement strategy to evaluate its adoption of best-practice anxiety care and the effectiveness of NAMASTE and the training module. Provider perceptions were assessed with an anonymous survey, and chart review was conducted to determine provider adherence to algorithmic care.

After evaluation of the program’s first training session, in 2015, materials were revised, and a second training session was conducted later in the academic year to ensure basic understanding about when to test for and how to treat anxiety disorders. The revised training emphasized the importance of treatment adherence (especially retesting with the GAD-7 to evaluate treatment effectiveness), the need to adjust medications, and strategies to optimize benzodiazepine management. Printed copies of the algorithms and medication lists were made available for resident precepting sessions, and training materials were available online.

A 2017 anonymous survey conducted to assess provider perceptions of NAMASTE received responses from residents (N=22) and attending physicians (N=18). Forty-one percent of residents agreed (N=9) and 18% strongly agreed (N=4) that training in NAMASTE enabled them to identify anxiety disorders, whereas 44% of attending physicians agreed (N=8) and 28% strongly agreed (N=5). After training in NAMASTE, 55% of residents agreed (N=12) and 18% strongly agreed (N=4) that their ability to treat anxiety disorders had improved, while 44% of attending physicians agreed (N=8) and 33% strongly agreed (N=6). Respondents reported that their ability to treat anxiety disorders with a scheduled medication had especially improved (residents, N=21, 95%; attending physicians, N=18, 100%). Residents were less likely than attending physicians to express confidence in their ability to choose augmenting medication alternatives to benzodiazepines (residents, N=12, 55%; attending physicians, N=11, 61%). Residents were also somewhat less likely to know when to refer patients for therapy; 50% of residents agreed (N=11) and 5% strongly agreed (N=1) that they had such knowledge, whereas 33% of attending physicians agreed (N=6) and 33% strongly agreed (N=6). Based on survey responses, training was revised with an emphasis on how to augment medications and optimize benzodiazepine management. After this training, resident physicians involved in quality improvement projects performed a chart review of GAD-7 retesting, medication adjustments, counseling referrals, adequate documentation, and medical follow-up intervals. Charts were reviewed for testing rates before and after training and implementation; fidelity to the algorithm, including GAD-7 retesting; follow-up physician appointments; medication prescriptions; counseling; and psychiatry referrals. To collect information related to program uptake, a variety of learner teams conducted several iterations of monthly resident-led quality review projects with convenience samples of patients. Continued chart review using different convenience samples indicated progress in achieving a standardized treatment approach, indicated by increased use of appropriate first-line medications and limited use or notations for contraindications of benzodiazepines with efforts to transition patients to alternative augmenting agents.

Conclusions

Overall, after being trained to use NAMASTE, clinicians indicated greater recognition of anxiety disorders in a primary care population and greater recognition of comorbid depression via retesting with GAD-7. Clinicians also indicated greater confidence in using algorithm- and measurement-informed clinical decision making to treat and help reduce the severity of anxiety disorders. Clinicians had a greater understanding of the role of anxiety contributors; self-care management; and when to initiate medication, behavioral counseling, or psychiatric consultation and assessment. They were also more aware of which medications were first-line for anxiety treatment and which may heighten anxiety initially or worsen anxiety overall. Provider uptake of NAMASTE has been high, and although further study is needed, we have identified indicators of patient improvement under algorithm-informed care.

Study of NAMASTE-guided care and training has focused on evaluating providers’ perceived knowledge and capacity and on evaluation of algorithm-consistent treatment efficacy. Repeated review is needed with consistent samples and sample criteria to more accurately assess treatment approach and patient outcomes. Plans to improve the training module include further emphasizing the importance of first- and second-line agents with safe and effective pharmacological options. We will also highlight when to refer patients to CBT, emphasize that titrating medications to therapeutic levels may enable greater uptake of the principles of CBT, and underscore the benefit of mind-body medicine for developing additional stress management skills. CBT and mind-body medicine training, which may be more acceptable to some patients than medication, may work better after a patient has been treated using medications, but this approach is controversial.

The IMC successfully integrated behavioral health care for depression into primary care. As the IMC and other examples of algorithm- and metric-driven depression and anxiety care have shown, a systematic method to identify and treat anxiety disorders in primary care can increase uptake among residents and attending physicians in a complex clinical setting and can improve patient health. Given the morbidity and mortality associated with untreated anxiety disorders, this method bears continued uptake, refinement, and study.