4

Exploring Innovative Approaches

The next panel session described innovative approaches to addressing the information and communication needs of those with mental health or behavioral health conditions. Two panelists, Joanne Nicholson, a professor in the Institute for Behavioral Health at Brandeis University, and Jessica Black, the director of Collaborative Care at NYC Health + Hospitals, spoke about innovations developed by medical care systems. Two additional panelists, Brandon Pate, a critical care paramedic with MedStar Mobile Healthcare, and Albert Park, a National Institutes of Health (NIH)–National Library of Medicine postdoctoral fellow in the Department of Biomedical Informatics at the University of Utah, then described community-based approaches. An open discussion was moderated by Teresa Wagner, a senior fellow for health literacy at the University of North Texas Health Science Center.

WEB-BASED RESOURCE FOR INDIVIDUALS WITH SERIOUS MENTAL ILLNESS¹

The goal of the Mental Health Bridges project, as Joanne Nicholson explained, is to provide usable and useful tools, resources, and opportunities for learning and practicing health literacy skills in an accessible online space with the aim of helping people living with serious mental illnesses overcome challenges in accessing and using health information and, ulti-

___________________

¹ This section is based on the presentation by Joanne Nicholson, a professor in the Institute for Behavioral Health at Brandeis University, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

mately, of reducing the health disparities that these individuals often face. Underlying this goal, Nicholson said, is the belief that health knowledge leads to informed decision making, which in turn leads to better management of health conditions and active involvement in health care. This is an important issue, she said, because people living with mental health conditions make up one of the largest health-disparate populations in the world and mental health conditions are responsible for 8 million deaths worldwide each year (Walker et al., 2015). Many of these deaths arise from preventable diseases commonly associated with smoking, obesity, and other factors prevalent among people with serious mental health conditions.

The majority of Americans get their health information online (Horrigan, 2016; Moreland et al., 2015), Nicholson said, adding that she does as well and noting that over the past year she has spent many more hours online looking for health information than she has with her primary care physician. The key question for her, she said, is how to translate online information into actual changes in health behavior, reduced disparities, and improved health outcomes for people with serious mental health conditions. While people with disabilities are less likely to go online and more likely to report being confused or challenged by using the Internet (Anderson and Perrin, 2017), Nicholson said, it is still the case that individuals with serious mental health conditions have cell phones and computers and use them daily. “Whether they are able to access and use the kinds of health information that we are hoping to promote is another question,” she said, “but it is no longer the case that we can believe that people with serious mental health conditions do not have smartphones or do not use the Internet.”

As was previously discussed, people with mental illnesses may have cognitive impairments that interfere with accessing information and putting it together to make informed decisions, particularly about health care or health behavior. There is now significant evidence, Nicholson said, that effective design of technology-based “spaces” and interventions can reduce cognitive burden (Naslund et al., 2017; Rotondi et al., 2015, 2017). For example, researchers found that a website designed using what is known as the flat explicit design model required less time to find information, had a higher success rate, and was rated easier to use and less frustrating than other websites by individuals with schizophrenia (Rotondi et al., 2015).

For the Mental Health Bridges project, Nicholson and her collaborators formed six focus groups across the country that included both individuals with a serious mental illness—all of whom reported using their cell phones and the Internet daily—and health information specialists; these specialists included medical school librarians, academic reference librarians, and clinicians who routinely provided health information. The focus groups were designed to find out what both types of people thought about health information and online access for people with mental health conditions. What

her team heard from the health information specialists was particularly interesting, Nicholson said. Before the advent of the Internet, health information specialists would often develop trusted relationships with people who came looking for health information. Today, however, much of their effort has shifted to teaching people how to find information online, and the time spent and consequently the quality of their relationships with their clients and patrons have been reduced.

One finding from these focus groups was that many times patients and professionals do not know exactly what they want to find out when they start searching for information; instead, when they look for information they start with a Google search and go from there, typically to the first or second listing on the search returns. The problem is that the first or second listing may not be a credible source of information and could even be an advertisement. Another common starting point is Wikipedia, where the consistent format for every entry helps people feel comfortable. Focus group members did report that they tend to stick to sites that have been helpful in the past and that they look for a name with authority, such as websites with government, educational, or nonprofit domain names. Professional organizations, such as the American Heart Association, are also common search destinations, and individuals seeking to get insights from others with similar symptoms or who are taking similar medications often visit disease-related social media sites. “For general information, they go to the professional sites, but for real day-to-day coping and management, they go to social media,” Nicholson said.

Nicholson offers a list of design suggestions for websites targeted at people living with serious mental illness:

• use vivid, warm colors and consistent formatting;
• include infographics;
• incorporate videos, audios, and images of “people who look like me” and role models;
• make navigation simple;
• present information chunks on the same screen;
• do not use too many columns;
• include tabs at the top of a page with drop-down menus;
• enable pages to scroll slowly;
• include options for multiple languages;
• add a search box; and
• make the web pages printable.

She also offered several suggestions regarding content. These included

• use clear, concise, and simple language written in a hopeful, positive tone;

• include practical suggestions on how to cope with daily challenges as well as information on common experiences;
• have a clearly defined title and scope for any materials;
• include an orientation to the site;
• be respectful and do not talk down to users, preach to them, or stigmatize them;
• brand the information, such as offered by a particular university;
• use local resources;
• include disclaimers; and
• avoid acronyms, clickbait sites, Google ads, pop-ups, and paid ads.

Nicholson concluded her presentation by noting that good design is not just about providing credible information in a form that individuals with a serious mental illness and their family members can use. Good design of digital resources has the potential to engage individuals living with serious mental illness with the resources, encourage them to make better choices and take charge of their health, and provide the structure and support to do something with the information in order to enhance their health and wellness.

HEALTH SYSTEM INTEGRATION OF MENTAL HEALTH INTO PRIMARY CARE²

NYC Health + Hospitals is the nation’s largest public hospital system, Jessica Black said. It serves more than 1.2 million New Yorkers annually at its 11 acute hospital sites, 5 long-term care facilities, and dozens of community health centers and school-based clinics. Some 53 percent of the patients it serves are Medicaid beneficiaries, and another 26 percent or so are uninsured, and the system’s clients speak more than 190 languages.

The reason it is so important to integrate behavioral health into primary care, Black said, is that between 30 and 50 percent of individuals referred from primary care to a behavioral health provider do not even make a first appointment at the behavioral health clinic, she said. In addition, fewer than 15 percent of patients with easy access to primary care who are accurately diagnosed with depression receive adequate treatment (Fisher and Ransom, 1997). Furthermore, Black said, two-thirds of primary care physicians report not being able to access outpatient behavioral health care for their patients as a result of shortages of mental health providers, health plan barriers, and lack of coverage or inadequate coverage.

___________________

² This section is based on the presentation by Jessica Black, the director of Collaborative Care at NYC Health + Hospitals, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

Difficulties with navigation and communication gaps between siloed primary care and behavioral health settings present another barrier to patients receiving coordinated, holistic physical and mental health care, Black said. Today, patients are often left to navigate various health care settings and coordinate their own care, including relaying information between providers. The goal of the Collaborative Care Program that Black heads is to create a “no wrong door” approach to quality health care, one in which patients can walk into a single point of entry and access all of the care they need. “From their perspective, it feels seamless,” Black said, describing the ideal situation.

Collaborative Care is based on the evidence-supported Improving Mood-Promoting Access to Collaborative Treatment (IMPACT) model, a collaborative, stepped-care approach in which a trained depression care manager—usually a nurse, social worker, or psychologist—works with the patient, the patient’s primary care provider, and a psychiatrist to develop and administer a course of treatment (Camacho et al., 2015; Grypma et al., 2006; Van Leeuwen Williams et al., 2009). This model has been tested in more than 80 randomized controlled trials and a meta-analysis shows that for depression, this model is consistently more effective than usual care. More recently, Black said, clinical trials have shown this approach to be effective in treating anxiety, posttraumatic stress disorder, and some pediatric patients.

Collaborative Care, Black said, improves access to mental health care by bringing patient-centered care to where the patient is and meeting patients where they are. “One of the things I love about Collaborative Care,” she said, “is it is really flexible in terms of how frequently the patient has to see their health care provider, allowing visits in person or over the phone, and it is holistic and treats the whole patient.” Black added that Collaborative Care also improves chronic disease control rates and reduces health care costs (Unutzer et al., 2008). At NYC Health + Hospitals, the program was funded initially by a New York State grant, which allowed the system to hire staff and roll out universal depression screening and integrated treatment across the entire system. Today the program is largely funded by Medicaid reimbursements, and the Centers for Medicare & Medicaid Services (CMS) began reimbursing for treatment of Medicare beneficiaries in January 2017, which Black said has been good news for the program’s sustainability. Recently, the program has been expanded to treat patients with anxiety disorders, substance use disorders, and other behavioral health conditions as well as to serve pediatric clinics at four facilities and one of the system’s women’s health clinics. Black said that 61 percent of the adults enrolled in the Collaborative Care program have experienced significant improvement in depression symptoms.

The Collaborative Care team, Black said, includes the primary care physician, a Collaborative Care clinician or care manager (either a regis-

tered nurse or a social worker), and a psychiatric consultant. The primary care physician maintains point responsibility for the patient’s care and oversees the treatment plan. The patient drives the treatment plan, she added. “They are the ones who are hopefully making the decisions about what their treatment plan is going to look like and how it is going to go.” Many consulting psychiatrists spend approximately 30 percent of their time on Collaborative Care, consulting on cases through weekly case meetings and through ad hoc appointments with the primary care physician and nurse or social worker; they do not see most patients physically (some patients will see the consulting psychiatrist for a one-time, face-to-face consultation). “They are critical in terms of building capacity in primary care for our physicians to feel more comfortable identifying and treating mental health conditions,” she said.

One of the core components of the program is universal screening for behavioral health conditions using standardized instruments, such as the Patient Health Questionnaire-9 for depression. Primary care clinics at NYC Health + Hospitals are not yet screening universally for anxiety and drug or alcohol use with the Generalized Anxiety Disorder 7-Item Scale, the Alcohol Use Disorders Identification Test, and the Drug Abuse Screening Test, but they do screen the Collaborative Care patients when they are referred. Once a patient has been identified, the care manager does outreach and engagement with the patient at least once per month by whatever means the patient finds most convenient. Treatment plans are evidence based and use motivational interviewing, behavioral activation, problem solving, and medication. New patients and those who do not improve are discussed at weekly case meetings by an interdisciplinary team of clinicians. Patients are tracked and monitored in a registry to drive outreach, psychiatric consultations, and changes in the treatment plan as needed. The registry, Black said, identifies patients who might be lost to follow-up or who have not been in touch for a while, as well as those who are not satisfied with their treatment.

The original IMPACT model was more rigid in its design, Black said, while Collaborative Care allows for more flexibility based on the implementation needs of a given site. Treatment in the program is based on measurable goals and outcomes—set according to patient desires—that are defined and tracked for each patient. Treatments, Black said, are adjusted based on a monthly assessment until clinical goals are achieved. She added that the collaboration and communication between the primary care physician and behavioral health clinician makes life easier for patients and lets them know they have a team taking care of them.

The Collaborative Care teams use several strategies to normalize and educate patients about depression, Black said. “Many of our patients come from backgrounds where depression is heavily stigmatized and where seek-

ing mental health care is stigmatized,” she said. “Some patients come from cultures where nobody even talks about depression, and they do not even know that it is a thing. They do not know that the symptoms they are experiencing are treatable.” To reach these individuals, primary care clinics have hung flyers talking about the symptoms of depression in their waiting rooms. These flyers alert the patient that there are people in the primary care clinic whom they can talk to while also educating patients and their families about the symptoms of depression. The posters were developed in multiple languages and for use in adult or pediatric clinics.

In addition, patients self-administer the depression screening instrument while in the waiting room before every primary care visit, which helps make depression screening seem as normal as taking a blood pressure reading. Staff are also taught strategies for asking about depression and for introducing the screening instrument in a way that is intended to increase the patient’s comfort level for sharing information; to normalize feeling “down” in response to life’s difficulties; to make the connection between feelings and physical health; and to convey hope that a treatment is available.

For patients identified as having a behavioral health problem, scripts are available for the nurse or social worker to introduce patients to the Collaborative Care model. With just a few sentences, the care manager is able to educate the patient about the symptoms of depression, connect physical and mental health, and offer the patient a variety of treatment options. Some facilities, Black said, have added depression support groups because they found that some patients are not interested in one-on-one treatment and prefer the social support that can develop in the group context.

Relapse prevention planning is an important piece of the Collaborative Care model, she said, and NYC Health + Hospitals would like to provide more training on it to Collaborative Care clinicians. Black explained that once a patient’s depression has improved significantly, the care team and patient start to develop a relapse prevention plan. Patients identify potential triggers, how their symptoms manifest themselves, the coping strategies they have available to them, and whom they can reach out to when they feel they have had a relapse. The patient is given a copy of the plan, and it is entered into the patient’s medical record for the care team to reference.

One area that remains a challenge for many Collaborative Care teams, Black said, is discussing medication treatment and adherence with patients. Currently, about half of the system’s enrolled patients include antidepressant medication in their treatment plans. “That is okay,” she said. “Counseling is effective and works well, but some patients may not get better with counseling alone, and we want to be able to talk to them about medication.” The goal, she added, is to make a collaborative decision with the patient about medication based on individual experiences and what will

work well for that patient. These conversations occur throughout treatment. Black added that her health system’s providers have found the Mayo Clinic’s depression medication decision aid cards, which outline the different side effects of specific anti-depression medication, can help patients choose the right medication for them.

MOBILE INTEGRATED HEALTH MANAGEMENT³

MedStar Mobile Healthcare is a government agency providing emergency and non-emergency ambulance service for more than 1 million residents of Fort Worth, Texas, and 14 surrounding cities, Brandon Pate said. Its 500-plus employees answer about 140,000 calls annually and operate on a budget of $44 million with no tax subsidy. In July 2009 MedStar developed and implemented the High Utilization Group community health program—now called Mobile Integrated Healthcare—to address ambulance “super users,” defined as those individuals who call for an ambulance 15 times or more in a 90-day period. Many of those first enrolled in the program were experiencing homelessness and had significant mental and behavioral health disorders. In the years since, Mobile Integrated Healthcare has evolved and expanded to include other programs, although the overlap with mental health and behavioral health disorders remains significant.

The basic strategy of Mobile Integrated Healthcare, Pate said, is to identify and address the social, economic, environmental, and health care–related gaps and to develop interventions and solutions for those gaps that are patient-centered, collaborative, and integrated with the other factions of the health care system. “We wanted to increase the primary care physician’s reach into their patients’ lives,” he explained, “so we become the eyes and the ears of the physician.” Upon enrolling a patient in the program, Pate and his colleagues conduct a gap analysis and needs assessment that serves as the starting point for an action plan for each patient. Some patients, for example, suffer from food insecurity or a lack of transportation. Others need a primary care physician and access to community mental health resources.

Pate noted that this approach coincides with the Triple Aim of improving the community’s health, improving patient satisfaction, and reducing health care expenditures. It also enables him and his colleagues, through twice-per-week home visits, to advocate on behalf of their patients, assist them in navigating the health system, and educate them on their disease’s

___________________

³ This section is based on the presentation by Brandon Pate, a critical care paramedic with MedStar Emergency Services, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

processes and medications. Over the course of 1 week, he said, the MedStar team spends several hours with these patients, working with them to identify and address goals. “As ambulance providers, we are uniquely positioned and poised for these kinds of programs,” he said. “We are accessible in every community. We are available 24/7. We are expected and respected and a welcome source of medical assessment and care in people’s homes throughout the community.”

He and his colleagues develop unique relationships with their patients and the community, Pate said, one that has afforded them unparalleled insights into the social, economic, and environmental factors that affect their clients’ health. “Based upon those findings and observations and by utilizing motivational interviewing, we work in collaboration with the patients to develop goals and to outline steps to achieve those goals,” he said. They then work collaboratively with other health care and social entities in the community to help each patient achieve those goals. The partnerships with community organizations have developed into a substantial network over the years that includes almost every hospital in its service area as well as insurance companies, physician groups, care management organizations, home health agencies, hospice agencies, and mental health agencies. These relationships are integral to the success of the patients, Pate said.

He noted that the hardest part of his job is resisting trying to fix all of a patient’s problems. “It can be easy to go into someone’s home and tell them they need to do this and that, but if they are not engaged and do not want it, it is not going to happen,” Pate said. Generally speaking, he said, the problem is that the majority of the program’s clients have been unserved or underserved and lack motivation to be engaged in their health care. To remedy this problem, every community health paramedic receives training on the use of motivational interviewing as a communication and interaction technique. This technique allows Pate and his colleagues to exercise influence and motivate the patients themselves to drive change and adopt improved health behaviors. “We teach our providers to acknowledge that our patients are autonomous, and they are ultimately in control of their health care and behaviors,” he said. “It is up to the patient to become engaged. It is up to the patient to change their health behaviors, but it is up to us as providers to encourage that change and that engagement.”

An equally important part of training helps staff learn how to listen to their patients, which, Pate said, can be hard for paramedics as they are accustomed to responding to chaotic emergency scenes that require them to be the authority figure. The training stresses the importance of asking open-ended questions, using reflection and summary statement to challenge patients to change their ways of thinking and promote positive attitudes toward improving their health, and working with patients to encourage them to embrace their autonomy. “We let them define that, and we do not

preach to them or lecture them,” Pate said. The end result, he added, is that patients become more engaged in their health care and their self-efficacy and health literacy improve.

The materials that Pate and his colleagues provide to patients include information, written at the fourth-grade level, specific to their disease process. These materials were authored in house and are peer reviewed. The program also encourages patients to attend disease-specific classes offered by community partners and even helps provide transportation to these classes.

Pate told the story of one recent patient who had been referred to the program by the county hospital. This patient had several admissions each for behavioral health, hyperglycemia, and heart failure. On his first visit, Pate did a medication inventory, asking this woman to gather all of the medications in her house. After 10 minutes, she shows up with a large box completely full of several hundred prescription medication bottles, none of which had been opened. She confessed that every time she went to the doctor or the hospital she was sent home with prescriptions that she had no idea how to take, so she decided it would be better to not take them. Pate’s response was to provide her with an organizing pill box and teach her how to use it, but when he asked her to explain to him how she was going to use it, she could not because she could not read the words on the pill box. Pate got creative and added visual aids and colors, and that did the trick.

The bottom line, he said, is that the program has resulted in improvement in patient outcomes across all categories measured, including mobility, self-care, the ability to perform usual activities, pain and discomfort, and anxiety or depression. A measure of overall health status improved by 40 percent, and there has been a 54.5 percent decrease in the number of ambulance trips to an emergency department. The latter has produced savings of $9.2 million, or more than $15,000 per enrolled patient. “Both of these outcome measures illustrate the importance of improving patient engagement, promoting health literacy, and addressing all of the needs of our patients, not just the medical needs,” Pate said in closing.

USING LARGE SOCIAL MEDIA DATA FOR UNDERSTANDING MENTAL HEALTH⁴

Albert Park explained that the goal of his work is to use various technologies to reach wider audiences and to obtain what relevant information individuals with behavioral health issues might not share with clinicians.

___________________

⁴ This section is based on the presentation by Albert Park, a National Institutes of Health–National Library of Medicine postdoctoral fellow in the Department of Biomedical Informatics at the University of Utah, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.

His work focuses on social media, which 77 percent of Americans use on a daily basis to connect socially and talk with others about various topics, including health (Perrin and Jiang, 2018). “If we can extract health-related conversations, we can learn about the questions and issues people might have,” Park said.

In one project, for example, he used natural language processing to extract e-cigarette–related conversations from the social media site Reddit and then used machine learning to identify the topics people were discussing. Many people, it turned out, were using social media to get health information, not all of which was accurate. Social media users, for example, believe that e-cigarettes can help people stop smoking and that using e-cigarettes is a fun experience (Park and Conway, 2017b). In another study, Park found that health information on e-cigarettes may be too difficult for many in the public to understand, with much of it written at the college reading level (Park et al., 2017).

Regarding mental health, Park said he has been trying to understand the long-term effects of participating in an online depression community, talking to like-minded individuals, and socializing in online environments (Park and Conway, 2017a). “The implications of long-term participation in an online health community, especially the depression community, can be quite interesting, because we know having someone to talk to and rely on at a time of difficulty can be very beneficial for you, but at the same time, your negative emotion can spread to others,” he said. In non–mental health communities, for example, patients with cancer, arthritis, and HIV/AIDS have been shown to improve on scores of depression (Griffiths et al., 2009a,b), anxiety (Hoybye et al., 2010; Setoyama et al., 2011), stress (Setoyama et al., 2011), and negative mood (Shaw et al., 2006) to become better able to cope with their illnesses (Mo and Coulson, 2012). However, a substantial amount of negative emotion exists in the depression community (Gkotsis et al., 2017), and emotional contagion theory suggests that these negative emotions might spread through the electronic community.

To explore the long-term effects of participating in online depression communities, Park turned to Reddit, a collection of more than 88,000 active discussion communities that in 2015 had more than 83 billion page views. Using a linguistic inquiry and word counting program, he compared the positive and negative emotional states of people in a depression community and in three other online health communities. The results showed that participating in the depression community improved the participants’ mood as expressed in the words they used in their comments and questions and that they improved more on some dimensions of mood than did the members of the other three groups. “Members were improving their psychosocial emotional states as they participated in the community,” Park said.

One thing that Park said he noticed consistently was that when people started new conversations, they typically said they were new to the community and stated their problems in a disorganized and confusing manner that clearly revealed their anger and frustration. At the same time, he said, the replies from other Reddit users were organized and coherent. As had already been mentioned, many mental health conditions can impair language ability through their effects on the brain’s front lobes. These findings were from small-sized, laboratory-based studies, and this new study gave Park the opportunity to see if these language impairments show up in social media conversations and if those impairments improved over time.

For the study, he selected three mental health communities focused on depression, bipolar, and schizophrenia along with three control communities. Park said that research has shown that the signs of language impairment include less lexical diversity and poor sentence structure, organization, and articulation. His analysis found that the neurotypical health communities had higher lexical diversity in their comments, while the depression, bipolar, and schizophrenia communities had more difficulty articulating their thoughts. Over time, the quality of the participants’ comments improved, becoming easier to read and having more lexical diversity. He concluded that weaknesses in written communication could be improved by participating in social media.

DISCUSSION

Gayle Mathe from the California Dental Association began the discussion by asking Pate to describe MedStar’s reimbursement model and how the company’s services are supported financially. Pate replied that ambulance transport services are reimbursed as a benefit by CMS. For the mobile integrated health care program, hospitals are paying for readmission avoidance, and several Blue Cross Blue Shield plans pay for care delivered to their high users. In addition, MedStar partners with home health agencies and hospice agencies to intercept their patients when they call 911 and receive payment for that service. In short, MedStar makes arrangements with various partners so that they save money because MedStar meets the needs of their patients more effectively.

Jennifer Dillaha commented that she read a book called Flourish that has a chapter describing how many people with chronic conditions suffer from a psychological condition called learned helplessness that leaves them unable to help themselves even though they potentially could. After reading this book, she said, she became interested in the idea that improving health literacy could be an antidote for learned helplessness, and she asked the panelists if they had seen evidence of this in their work. Black replied that one of the key interventions her program’s nurses and social workers use

is behavioral activation, which involves scheduling pleasant activities. The idea is that people with depression have trouble activating themselves to do things they like to do even though they might feel better after engaging in that activity. Giving these individuals achievable goals involving being able to do those activities can help the cycle of learned helplessness that often accompanies depression.

Earnestine Willis from the Medical College of Wisconsin asked the panelists to describe how the information they gather about the social determinants of health and the level of health literacy of their patients gets back to the patients’ medical homes or primary care providers. Nicholson said that she and her collaborators are still in the process of building the Mental Health Bridges program, but it will include a process for bringing the health care plan developed with the patient back to the primary care provider. Pate said that his team is in constant communication with its clients’ outpatient case managers and that all of the documentation he and his colleagues develop either gets faxed to the referral source and primary care physician or uploaded to a patient’s EHR.

Black explained that because her program is embedded in the primary care setting, the nurse or social worker who is physically located in the primary care clinic is in close contact with the primary care provider and is able to enter information into the patient’s EHR. In addition, the program has what it calls a champion primary care provider who attends the weekly case consultation meetings, and the psychiatrist or a Collaborative Care nurse or social worker contacts the patient’s primary care provider to relay the results of those consultations. She noted that the program does not yet have any Collaborative Care clinics at any of its school-based health centers, but if they are established, any information gathered about treatment will likely be documented in the student’s EHR.

Lawrence Smith commented that his institution, Northwell Health, has an extensive program of embedding behavioral health workers—either social workers trained in psychiatric counseling or behavioral health nurse practitioners—in primary care offices that are both amenable and large enough, and he said that he was intrigued that Black’s Collaborative Care program used registered nurses, who are less expensive, to deliver care. He asked Black to discuss how her program trains its registered nurses and how it monitors whether they are delivering the specific therapy prescribed. Black said that the program decided to train nurses, a decision made before she arrived, because NYC Health + Hospitals was focused on the comorbidities associated with depression in order to make it easier for the patients to work with one person on both their chronic disease and their depression.

Today, she said, her program has Collaborative Care Clinical Coaches, people with clinical backgrounds who are in the facilities and work closely with the program’s nurses and social workers to provide training and in-

the-moment coaching. In addition, the consulting psychiatrists work with the nurses, social workers, and primary care physicians on improving their ability to work with patients with mental health issues and their comfort level in doing so. “I will say that it takes a special kind of nurse, even a special kind of social worker, to do this work,” Black said. “We do have high turnover in those roles, but for the people who want to do it and are interested, they stay and are really good at it.” She also noted that after 5 years of having psychiatrists working with the primary care physicians, most primary care physicians have now become comfortable writing prescriptions for antidepressants. Currently, the program is going through the same education process to get primary care providers comfortable prescribing anti-anxiety medications.

Ellen Markman from Stanford University asked Park to reconcile the idea that depression is contagious with the fact that interacting on social media with others who have depression appears to improve depression symptoms. Park said he believes that socializing with others in an environment designed to offer emotional support contributes to the positive benefits these individuals are experiencing. He noted that Reddit communities have a moderator and said that he sees evidence that they are removing at least some potentially harmful comments from the discussion boards. He has also seen the communities moderate themselves and let those who post potentially harmful comments know that that is not the purpose of these communities.

Olayinka Shiyanbola asked Park about the use of negative words such as “death” and “suicide,” and he said that the use of these words decreases as a user’s comments become more positive. Shiyanbola then asked him if there is an opportunity to monitor the conversations to spot an individual contemplating suicide and get that person help. Park said it might be possible, but he is not aware of it happening currently. As a final comment, O’Leary asked whether there should be a discussion about recommending that people participate in online communities, how to ensure that there are safe spaces in those communities, and whether providers might join those communities.

REFERENCES

Anderson, M., and A. Perrin. 2017. Disabled Americans are less likely to use technology. In FactTank: News in the numbers. Washington, DC: Pew Research Center.

Camacho, A., P. Gonzalez, S. F. Castaneda, A. Simmons, C. Buelna, H. Lemus, and G. A. Talavera. 2015. Improvement in depressive symptoms among Hispanics/Latinos receiving a culturally tailored impact and problem-solving intervention in a community health center. Community Mental Health Journal 51(4):385–392.

Fisher, L., and D. C. Ransom. 1997. Developing a strategy for managing behavioral health care within the context of primary care. Archives of Family Medicine 6(4):324–333.

Gkotsis, G., A. Oellrich, S. Velupillai, M. Liakata, T. J. Hubbard, R. J. Dobson, and R. Dutta. 2017. Characterisation of mental health conditions in social media using informed deep learning. Scientific Reports 7:45141.

Griffiths, K. M., A. L. Calear, and M. Banfield. 2009a. Systematic review on internet support groups (ISGs) and depression (1): Do ISGs reduce depressive symptoms? Journal of Medical Internet Research 11(3):e40.

Griffiths, K. M., A. L. Calear, M. Banfield, and A. Tam. 2009b. Systematic review on internet support groups (ISGs) and depression (2): What is known about depression ISGs? Journal of Medical Internet Research 11(3):e41.

Grypma, L., R. Haverkamp, S. Little, and J. Unutzer. 2006. Taking an evidence-based model of depression care from research to practice: Making lemonade out of depression. General Hospital Psychiatry 28(2):101–107.

Horrigan, J. B. 2016. Lifelong learning and technology. Washington, DC: Pew Research Center.

Hoybye, M. T., S. O. Dalton, J. Christensen, L. Ross, K. G. Kuhn, and C. Johansen. 2010. Social and psychological determinants of participation in internet-based cancer support groups. Supportive Care in Cancer 18(5):553–560.

Mo, P. K., and N. S. Coulson. 2012. Developing a model for online support group use, empowering processes and psychosocial outcomes for individuals living with HIV/AIDS. Psychology & Health 27(4):445–459.

Moreland, J., T. L. French, and G. P. Cumming. 2015. The prevalence of online health information seeking among patients in Scotland: A cross-sectional exploratory study. JMIR Research Protocols 4(3):e85.

Naslund, J. A., K. A. Aschbrenner, S. J. Kim, G. J. McHugo, J. Unutzer, S. J. Bartels, and L. A. Marsch. 2017. Health behavior models for informing digital technology interventions for individuals with mental illness. Psychiatric Rehabilitation Journal 40(3):325–335.

Park, A., and M. Conway. 2017a. Longitudinal changes in psychological states in online health community members: Understanding the long-term effects of participating in an online depression community. Journal of Medical Internet Research 19(3):e71.

Park, A., and M. Conway. 2017b. Tracking health related discussions on Reddit for public health applications. AMIA Annual Symposium Proceedings 2017:1362–1371.

Park, A., S. H. Zhu, and M. Conway. 2017. The readability of electronic cigarette health information and advice: A quantitative analysis of web-based information. JMIR Public Health Surveillance 3(1):e1.

Perrin, A., and J. Jang. 2018. About a quarter of U.S. adults say they are “almost constantly” online. Fact Tank: News in the Numbers. Pew Research Center. http://www.pewresearch.org/fact-tank/2018/03/14/about-a-quarter-of-americans-report-going-online-almostconstantly (accessed February 7, 2019).

Rotondi, A. J., S. M. Eack, B. H. Hanusa, M. B. Spring, and G. L. Haas. 2015. Critical design elements of e-health applications for users with severe mental illness: Singular focus, simple architecture, prominent contents, explicit navigation, and inclusive hyperlinks. Schizophrenia Bulletin 41(2):440–448.

Rotondi, A. J., M. R. Spring, B. H. Hanusa, S. M. Eack, and G. L. Haas. 2017. Designing ehealth applications to reduce cognitive effort for persons with severe mental illness: Page complexity, navigation simplicity, and comprehensibility. JMIR Human Factors 4(1):e1.

Setoyama, Y., Y. Yamazaki, and K. Namayama. 2011. Benefits of peer support in online Japanese breast cancer communities: Differences between lurkers and posters. Journal of Medical Internet Research 13(4):e122.

Shaw, B. R., R. Hawkins, F. McTavish, S. Pingree, and D. H. Gustafson. 2006. Effects of insightful disclosure within computer mediated support groups on women with breast cancer. Health Communication 19(2):133–142.

Unutzer, J., W. J. Katon, M. Y. Fan, M. C. Schoenbaum, E. H. Lin, R. D. Della Penna, and D. Powers. 2008. Long-term cost effects of collaborative care for late-life depression. American Journal of Managed Care 14(2):95–100.

Van Leeuwen Williams, E., J. Unutzer, S. Lee, and P. H. Noel. 2009. Collaborative depression care for the old-old: Findings from the impact trial. American Journal of Geriatric Psychiatry 17(12):1040–1049.

Walker, E. R., R. E. McGee, and B. G. Druss. 2015. Mortality in mental disorders and global disease burden implications: A systematic review and meta-analysis. JAMA Psychiatry 72(4):334–341.