Navigating the ethics of Internet-guided self-help interventions

Abstract

Internet-guided self-help (IGSH) programs have proliferated recently to treat common mental health problems such as anxiety and depression. However, technology has outpaced the development of ethical guidelines for this mode of delivery. We examine ethical challenges in this new space, including defining the role “guides” play in treatment, crisis management, and user selection and screening. IGSH programs can provide safe and ethical care when they (a) coordinate care effectively with other systems; (b) provide competent and well-defined guidance; and (c) reach users that are appropriate for and well-educated about the services they are going to receive. We argue that jurisdictional practice constraints and outdated regulatory and ethical guidelines may impede the ability of IGSH programs to maintain or even improve performance when faced with greater demand, larger populations, heterogeneous settings, and the desire for large-scale dissemination.

1 |. INTRODUCTION

Over the past decade, mobile and Web technologies have infiltrated every part of our lives, changing the way we communicate, learn, entertain ourselves, monitor our health, and manage our personal and professional lives. Inevitably, a wide variety of products have emerged with an aim to use technology to manage and improve our mental health. Regulatory guidelines have thus far primarily focused on the provision of “traditional” telemedicine services; this generally includes psychotherapy, assessments, consultation, or psychiatric services provided via videoconferencing or telephone (Maheu, Pulier, McMenamin, & Posen, 2012). Little attention has been given to developing standards for online programs that provide psychosocial treatment, coaching, or other support direct-to-consumers’ mobile phones or computers—an increasingly common mode of delivery.

One burgeoning form of online service is Internet-guided self-help (IGSH) programs, defined for the purposes of this article as any mental health program delivered via mobile or Web that is primarily self-guided, and also includes facilitation or support provided by a professional or paraprofessional. Providers offering IGSH are bound to the same ethical codes as their in-person counterparts; however, the application of ethics standards in an online, guided self-help format can create unique challenges. This article ventures to illuminate ethical issues that arise in the implementation of IGSH programs, and the practical implications that must be addressed to deliver programs that maximize benefit and minimize the potential for harm to consumers.

2 |. IGSH WITHIN THE TREATMENT LANDSCAPE

Internet-guided self-help programs represent a relatively new development within the broad context of mental health treatment modalities. It is important to differentiate IGSH from related approaches and to understand how ethical guidelines and issues related to other treatment modalities may or may not extend to IGSH. Much like traditional guided self-help treatments, IGSH may be considered along with a spectrum of support between pure self-help (traditional or Internet-based) and individual psychotherapy including telehealth. In fact, the American Psychological Association’s (2013) Guidelines for the Practice of Telepsychology explicitly acknowledge Web-based self-help as falling under the broader telepsychology umbrella, though they are largely silent on IGSH specifically. Existing in a space between self-help and therapy, IGSH programs pose unique challenges and highlight the need for professional ethics to evolve alongside technology.

The rise of IGSH models is an exciting development within the broader telemedicine field, given IGSH programs’ potential for widespread distribution, scalability, and a growing evidence base. As compared to one-on-one telehealth services, IGSH programs typically are designed to serve a significantly larger number of users per “guide”¹ without compromising clinical effectiveness or user satisfaction. The scale of IGSH programs offers the potential to drive down costs for organizations and consumers, reduce wait times for care, and extend the reach of empirically supported treatments. IGSH programs are typically based on cognitive behavioral and related therapies, although preliminary research suggests that psychodynamic interventions viably may be adapted to IGSH as well (Andersson et al., 2012). Research on the efficacy of IGSH programs has targeted a wide variety of mental health conditions such as anxiety disorders, (Hadjistavropoulos et al., 2014; Schneider, Mataix-Cols, Marks, & Bachofen, 2005), depression (Andersson & Cuijpers, 2009; Richards et al., 2015), and disordered eating, including binge eating disorder and bulimia nervosa (Carrard et al., 2011; Wagner et al., 2013). IGSH programs frequently target individuals with subthreshold or mild-to-moderate clinical presentations, although there is some evidence that such programs may treat severe mental health symptoms effectively as well (Bower et al., 2013).

Internet-guided self-help program offerings vary in terms of content, duration, frequency and modality of contact, and the role that guides play in supporting the individual’s learning. They are short-term interventions; in the research we reviewed, it was most common for programs to comprise 5–10 weekly modules. Support is typically provided via telephone or online on a weekly basis, is initiated by the therapist or in response to clients’ questions/expressed concerns, and is brief in nature (e.g., 10–20 min; Drozd et al., 2016). Guidance is provided most commonly by professional mental health therapists, but may also have been delivered by advanced graduate students, primary care physicians, nurses, and peer counselors (Day, McGrath, & Wojtowicz, 2013). Available evidence suggests that the professional credential of the guide does not have a significant effect on the success of an IGSH intervention. For a more thorough review of the impact of guidance on digital interventions, see Brown and Jones (2017).

Despite a growing research base demonstrating the efficacy of IGSH, these programs are not readily available in routine care in the USA, arguably due to the lack of training available to providers, lack of access to programs, lack of attention to organizational drivers for implementation, and concerns about competition with face-to-face care providers (Drozd et al., 2016; Wyatt & Sullivan, 2005). In contrast, the UK, for example, includes guided self-help including computerized CBT as a recommendation for subthreshold or mild-to-moderate depression, panic, and generalized anxiety disorder within their National Institute for Health and Care Excellence guidelines for a stepped care model of treatment (National institute for Health and Care Excellence, 2011). IGSH has also been introduced via workplace health and wellness programs to address common mental health concerns and improve lifestyle (e.g., reduce problem drinking, and improve sleep), which in turn may reduce absenteeism and improve productivity (Joyce et al., 2016). The growing adoption of IGSH programs underscores the need for regulatory guidance and oversight in their implementation.

3 |. ETHICAL CONSIDERATIONS IN DEVELOPING AN IGSH PROGRAM

This article attempts to examine some of the challenges in the development and delivery of ethically sound IGSH and propose suggestions for future consideration. To do so, we focused on three broad questions: How can ethics inform how IGSH applications design their programs and frame the role of their guides? What are the challenges associated with managing clinical risk in the context of these platforms, and how can we minimize the potential for negative impacts? Lastly, how can we ensure that our user base is appropriate for and well-informed about the risks and benefits of IGSH? The suggestions we arrive at—and in many cases, the additional questions that emerge—are intended to be aspirational in nature and focused on developing best practices, rather than more narrowly focused on the question of avoiding litigation or meeting specific legal requirements (which are, in any case, still murky in this emerging space).

4 |. DELIVERING AND FRAMING AN IGSH PROGRAM

4.1 |. Model

Internet-guided self-help programs are typically deployed in one of the two models: stepped care and centralized units. Stepped care interventions are offered to patients on a “least restrictive” basis, generally meaning treatment is initiated at the lowest intensity for which a positive response can be expected. Group therapy and guided IGSH programs typically represent a “second step” following pure self-help. Higher levels would include in-person, brief- or longer-term therapy and inpatient treatments (Newman, 2000). The impacts of less restrictive steps are monitored systematically, and care is stepped up as necessary (Bower & Gilbody, 2005). In stepped care models, IGSH is typically delivered within the primary care clinic by a trained physician, social worker, psychiatric nurse, or other embedded mental health practitioner who retains responsibility for the patient’s care, and will monitor progress to determine whether the patient can be considered discharged successfully, or requires additional treatment. The UK’s nationalized health care system has adopted a stepped care model of treatment and utilizes IGSH interventions to provide less restrictive mental health treatment to their patients within the primary care system.

In centralized unit models, a stand-alone entity is responsible for all aspects of the intervention, including developing the intervention; maintaining the software involved; training and monitoring therapists or guides; and screening, assigning, and monitoring patients. For example, in one Canadian study, the “Online Therapy Unit for Service, Education, and Research” effectively coordinated computerized CBT for adults with depression and anxiety across multiple sites, using trained community therapists as guides (Hadjistavropoulos et al., 2014). A study in the UK delivered a computerized self-help program for anxiety and depression, paired with brief telephonic or in-person support, via primary care referral to a stand-alone CCBT clinic (Marks et al., 2003).

From an ethical perspective, benefits and risks for patients exist for both models. A stepped care model provides strong continuity of care, as the designated practitioner refers clients to IGSH and monitors treatment response to “step up” clients’ care as needed. Providers can also help connect patients to appropriate local resources in case of an emerging risk situation, such as disclosure of suicidal thoughts or serious self-harm. In contrast, a main advantage of the centralized unit model is that it offers the ability to maintain a consistent standard of service across guides (Drozd et al., 2016). However, providers in the centralized unit model have no previous relationship with patients, are likely to have less access to comprehensive clinical information, and have limited tools available to support their treatment. Given these limitations, it is especially important for centralized units to develop robust risk management policies and procedures. It is worth examining the merits of hybridizing these models to attain benefits inherent to each—for example, by centralizing an IGSH unit while building routine coordination with primary care into their operations.

4.2 |. Role of guidance

Broadly speaking, the role of guidance in IGSH is to provide a supportive and nonjudgmental environment, encouragement, and promote adherence to treatment through the provision of structure and accountability (Brown & Jones, 2017). Guides also provide clarification about any misunderstandings of the techniques provided (Newman, Erickson, Przeworski, & Dzus, 2003). Means of communication with users may include synchronous (simultaneous) or asynchronous (delayed) communication, a set or a flexible communication schedule, and a variety of media such as phone, email, text, or in-app messaging. Guidance is distinct from psychotherapy in that the online intervention delivers the bulk of the therapeutic content, with the guide serving a secondary support function. Further, whereas a therapist may provide 12–16 hr of face-to-face contact to deliver one round of CBT, it is not uncommon for guides to spend 5–10 min/week supporting and responding to an individual (Newman, Szkodny, Llera, & Przeworski, 2011a, 2011b), which could translate into <1 or 2 hr total time spent on the average user.

The scope of a guide’s practice is significantly more limited than that of a therapist’s, although the distinction between guidance and therapy may sometimes be challenging to define. For example, if guides are tasked with helping to motivate users, positive reinforcement and reminders fall clearly within this job description. However, consider the example of a user with a generalized anxiety disorder who states she is struggling to engage with a CBT-based IGSH program due to “lack of family support.” Helping her to problem solve may enable her to effectively engage with and benefit from the intervention, but may also require an additional assessment to better understand the family culture, history, and dynamics. Whereas supporting the user in applying problem solving to her situation would likely fall within the scope of practice for a guide, a therapeutic discussion on family history and culture would likely fall within the scope of psychotherapy and may be inappropriate for an IGSH program.

Distinguishing the scope of practice for guidance versus psychotherapy can be especially important when licensed mental health professionals are engaged as guides. Regulators are more likely to claim jurisdiction over services provided to address clinical issues similar to those traditionally presented in therapy, and those that use psychologically robust methods, regardless of whether the service claims to be providing psychotherapy (Harris, 2009). Indeed, direct-to-consumer services often include statements in their terms of service specifying that they are not a substitute for treatment and do not constitute medical or psychological advice. Despite this, regulatory boards may decide that this does not pass the “duck test,” a possibility that is as yet untested legally. As IGSH becomes more common, eventual litigation of these questions seems inevitable, particularly around issues of protecting clients from harm, mandated reporting, and the provision of therapeutic services across state lines (Association of State and Provincial Psychology Boards, n.d.).

Such concerns pose a challenge for IGSH providers, given potentially divergent interpretations of the nature of guidance and the applicable regulations among jurisdictions. The APA guidelines for practice of telepsychology state that psychologists are responsible for knowing and abiding by requirements of the location in which they are situated, as well as requirements of the states in which their clients are located (American Psychological Association, 2013). Requirements for licensed practitioners can vary dramatically across discipline and from state to state. For example, in Georgia, licensed counselors and social workers must complete a 6-hr course on telehealth competencies before legally engaging in services provided via Internet, telephone, smartphone, or other electronic media; it requires an additional 3 hr of training for telehealth supervisors (Maheu, 2015). Arkansas similarly requires mental health professionals to complete a technology-assisted counseling specialization certification before delivering any form of remote mental health services, whereas Connecticut requires all behavioral health professionals to have access to the patient’s medical history including the name and address of the patient’s primary care physician before initiating services remotely, including IGSH (State Senate Bill No. 467). On the other hand, if clients in these states are not under the care of a behavioral health professional or in some cases, a licensed health professional, IGSH remote services may be delivered to them with no oversight or stipulations in place.

In addition to defining the scope of practice for guides, including specific task and role boundaries, IGSH programs must determine the appropriate backgrounds and credentials for those providing guidance. Available evidence does not demonstrate a significant improvement in outcomes when guidance is provided by a mental health professional as compared to nonprofessionals, suggesting that nonspecialists can effectively provide this type of support (Brown & Jones, 2017). Unlicensed guides, such as health educators, coaches, or trained laypersons, do not fall under the authority of regulatory boards and could potentially avoid legal complications while being less expensive for IGSH businesses to employ. However, questions remain as to what level and type of supervision and professional oversight should be available to paraprofessional guides to ensure they can adequately support clinically complex or higher risk users. When paraprofessional guides are utilized, they should be supported to work within their scope of competence, and depending on the model, should have access to internal or external resources to refer and escalate higher risk users to the appropriate professionals.

4.3 |. User selection and screening

A third consideration in developing an IGSH program is defining the target population and deciding how prospective clients are screened and admitted. Typically, guided self-help is considered most appropriate for patients with mild-to-moderate symptoms, given the likely presence of clinical complexity and risk management concerns with more severe presentations (National Institute for Health and Care Excellence, 2011). Users are often referred to IGSH via physicians or other healthcare providers, who may conduct their own screening or assessment to determine whether this level of care is appropriate. Where IGSH is integrated into the care system, this process may be formalized, although outside of such systems providers may simply use their best clinical judgment to refer patients to options based on familiarity, reputation, and availability of resources. In a direct-to-consumer model, prospective users self-identify as likely to benefit from an intervention based on their own self-assessment of the problem, and their own understanding of the IGSH program (usually based on marketing materials). IGSH may also be offered to members of an organization, for example, an employer or school wellness benefit, who then follow a similar path of self-identification and sign up. Prospective users may sign up directly or may be subject to additional screening to determine whether the program is an appropriate fit.

Who are appropriate clients for IGSH? Symptom severity is often cited as an important consideration, with severely impaired or high-risk users typically being screened into more intensive treatment options. However, there is some evidence that higher severity users may also benefit from IGSH interventions. Noting the high percentage (40%) of interested patients that were screened out of an IGSH trial due to exclusion criteria, Hadjistavropoulos et al. (2014) suggested the potential for reaching more patients by reducing exclusion criteria, particularly when coordination of care is available to quickly “step up” users as necessary.

Although IGSH is generally considered a low-risk intervention, an “open to all” policy is subject to potential pitfalls, and inclusion/exclusion criteria should be considered thoughtfully. Patient preference should be one factor in determining whether a technology-based intervention is appropriate, but it should not be the only one (American Psychological Association, 2013) as users may self-select into inappropriate lower-level treatment options due to financial and time constraints, lack of information, stigma, or avoidance. In the authors’ experience, providing IGSH programs at a company that provides mobile CBT for anxiety, depression, and disordered eating, allowing for open enrollment into these tracks has at times created unintended consequences. For example, some users have attempted to use exposure therapy modules for inappropriate purposes, such as attempting to increase time spent with a known abuser or to address complex trauma without first establishing adequate coping skills or external support. In other cases, poorly matched users have had a negative first experience with mental health services, potentially decreasing the likelihood that they will seek other empirically supported treatments in the future. On the other hand, it is possible that IGSH may in many cases be a “better than nothing” option for users who would otherwise avoid treatment altogether. This is an important consideration, given that increasing access to care is central to the mission of digital health. For some users, IGSH may also serve as a bridge to other treatment options (White, 1998, 1995).

How do IGSH providers balance accessibility of the service with avoiding negative consequences to less-appropriate users (and the potential of legal risk)? We believe that three main factors should be considered. First, is there empirical support for the use of a given intervention with the targeted population? This may include consideration of clinical, demographic, and cultural factors. Second, are prospective users appropriately informed of the scope and boundaries of the service? Marketing claims should always align with the research, and an informed consent process ensures user understanding. Mental health professionals have long expressed concerns regarding the exponential increase in self-help books and other materials, which are often produced and disseminated without established evidence for their efficacy or appropriateness. Indeed, some psychologists and counselors have called for clearer ethical guidelines regarding professional responsibilities in the selection of self-help books for use with clients. IGSH programs face the same challenges, and arguably raise a more pressing need for regulation given that these programs can be developed and disseminated much more quickly than printed books. For example, it might be prudent to require an age limit for consent to participate and/or parental consent for those below a certain age to ensure personal vetting of content. The American Psychological Association or other organizations could also develop a regulatory body that reviews and provides a stamp of approval for high-quality self-help and guided self-help options. The journal Cognitive and Behavioral Practice which is affiliated with the Association for Behavioral and Cognitive Therapies has begun a new initiative to provide reviews of such interventions to help therapists know which ones they should recommend. Finally, there must be protocols in place to identify and support users who are unlikely to benefit from IGSH. This can include guidelines for referral or screening measures used prior to signup, as well as protocols that guides may use to redirect users’ behavior within a program, refer them to alternative resources, or escalate them to crisis support as necessary.

Other considerations for prospective IGSH clients include their technology fluency, reading and comprehension ability in the language of the intervention, availability of accommodations for a patient’s disability status, and cultural appropriateness of the program (American Psychological Association, 2013). For example, a user who struggles with reading and writing in English (a second language) may find it challenging to process information being presented and may also struggle to communicate these challenges to a coach. In a worst-case scenario, concerning risk factors may go overlooked due to communication challenges. Likewise, users who are new to using app-based technology may experience frustration, lack of motivation, and hopelessness in their struggle to work through a mobile program, and may also be challenged to communicate these struggles to a guide through the Web app. These issues are nontrivial as they can disrupt users’ core ability to benefit from an IGSH program. In fact, the state of Delaware requires assessment of technology fluency and the need for disability accommodations as part of their requirements for prescribing a telehealth solution (Epstein Becker Green, 2016).

The APA practice guidelines for telepsychology require that psychologists adhere to the same standard of care when providing services online, and furthermore suggests that psychologists should take extra care in assessing prospective clients for remote services given their relative newness and rapid evolution (American Psychological Association, 2013). In practice, assessments to determine user appropriateness for a given IGSH intervention may range from a thorough, in-person, or video-based diagnostic interview with a licensed practitioner, to a brief screening questionnaire offered directly through the IGSH application itself. Symptom inventories and risk-screener questions can easily be deployed in the latter scenario, although a professional assessment is better equipped to identify issues such as a cognitive disability or cultural mismatch between an IGSH intervention and the prospective user’s needs. Again, this trade-off illustrates the tension between the robustness of assessment in the referral process and the accessibility and scalability of IGSH services.

4.4 |. Takeaways

The way an IGSH program chooses to define its scope—including decisions around its chosen model, role of guidance, and target audience—sets the stage for ethical practice and will have a profound impact on how clinical risk is managed as discussed below. In this emerging practice area, we look to researchers and our ethical and regulatory bodies to investigate and promote empirically supported best practices. In particular, the resolution of jurisdictional issues is a high priority. In addition, professional organizations must update their telehealth practice standards to adapt to changing technologies. A recurring theme in ethical decision making in the IGSH format is navigating the tension between supporting the scalability and accessibility of online services, and the robustness of safeguards against unintended negative consequences.

5 |. MANAGING CLINICAL RISK

Whenever a mental health service is offered, a proportion of distressed individuals can be expected to present with some risk of danger to themselves or others. Risk may be imminent, such as users disclosing their intention to take their own lives, or can be chronic, in the case of a user with ongoing passive suicidal ideation, feelings of hopelessness, and recurrent struggles with substance abuse. Other risk factors may be external, such as users who disclose that they are in physically abusive relationships. Managing risk is both an ethical issue and a legal issue to the organization offering IGSH and the guide; however, being at the frontlines of the intervention, the guide is the person most likely to interact with and make case-by-case decisions around risk management.

The model and framework of the IGSH program have a significant impact on the resources available to guides as they manage risk. One decision point to consider is the provision of anonymous services versus verification of users’ identities. Given the ongoing stigma toward mental health services, anonymity continues to hold a significant draw for individuals who may be unwilling or unable to seek other treatment. Anonymity is also congruent with many people’s expectations for receiving support online more generally (e.g., via peer support forums). On the other hand, the provision of anonymous services has created ethical and practical concerns for online practitioners; an issue explored not only in professional spaces (Ferguson, 2016; Maheu, 2011; Novotney, 2017) but also as a matter of urgent concern in the popular media (Ferguson, 2016). Anonymity can present serious challenges when users are in crisis, endorsing a serious danger to themselves or others, or sharing information that would trigger a mandated report. No practitioner or organization wants to be in the position of being unable to intervene when a user discloses imminent suicidality or discloses a child is involved in ongoing abuse. In addition to the consequences to IGSH users, in most states, failure by a mandated reporter to report abuse can result in civil and criminal penalties, with fines up to $10,000 and/or a jail sentence of up to 5 years.

At this time, offering users some degree of anonymity is a common practice for online mental health services. Even when no other information is collected about a user, IP addresses can often be obtained for the purpose of attempting to initiate a police welfare check. However, these can be unreliable sources of location data (e.g., if a user is logging in through a VPN). IGSH providers may also offer anonymity by creating a firewall between users and guides, while collecting information that may be deployed in an emergency “break the glass” situation. When companies collect billing information directly from users, they are likely to have access to a full name, address, and email address. In our experience, there is a great deal of variability in child protection agencies’ ability and willingness to receive reports from IGSH providers with limited identifying information. In one California county, for example, this author was able to file a Child Protective Services report providing only a first name and IP address; in another county, a report of a young child being hit by a parent was declined on the basis of insufficient identifying information, despite providing a full name and email address indicating a specific university affiliation.

The choice of synchronous versus asynchronous services also impacts how risk is managed. For example, in a real-time conversation, an ambiguous statement such as “I’m so humiliated I feel like dying” would likely trigger a clarifying question to ascertain whether the user was experiencing suicidal ideation or expressing herself hyperbolically. However, in an asynchronous conversation, the guide may not read such a message for hours or even days and may wait even longer to receive a response to a follow-up question. This is clearly not a viable time frame for conducting a risk assessment in a potential crisis. Therefore, it is important that participants in asynchronous IGSH programs are made aware of resources they can contact in case of a crisis situation—for example, their physician’s on-call service or designated crisis hotlines on their health plan. Outside of a coordinated care system, an alternative is to develop a protocol for when and how guides (or other representatives of the IGSH programs) can engage clients in a real-time conversation. For example, a designated crisis coordinator could be available on-call to proactively out-reach and respond to risk. A final option is to refer users to a general crisis hotline for additional assessment, such as a national or local suicide hotline. This option is free and widely accessible, although there is no way to ensure or verify that users receive the guide’s referral message in a timely manner.

To illustrate some of the issues at hand, consider the hypothetical case of a user who signed up for a program advertised online, which offers a subscription-based stress management program with guidance provided by a licensed, master-level clinician. This user discloses in his first message to the guide that he is currently seeking support around work stress. Over the course of the next few weeks, it emerges that the user has been previously diagnosed with bipolar disorder, and although he has been stable for the past year, his mood has been low recently. A month into the IGSH program, the user reports that he has discontinued medication, and his messages include some bizarre and ambiguous statements related to death and dying. The guide sends a message referring the user back to his physician and urges him to call 911 or walk into the ER if he is experiencing suicidal thoughts. The user, however, does not appear to have viewed the message and has had no further contact with the guide.

In this scenario, what responsibility does the guide (and more broadly, the IGSH program) have for the safety and well-being of this user? The user has not indicated a specific and imminent threat to himself that would very clearly indicate the need for police intervention, but the content of his messages and mental health history clearly indicate the possibility of a current crisis. In the therapist role, online or offline, responsible clinicians would likely make multiple attempts to follow-up on the dropout of a decompensating client with multiple known risk factors. Therapists may attempt contact via various means of communication, such as secure email, telephone, or even snail mail. They would also likely have emergency contact information and/or a release to speak with the prescribing physician, enabling them to reach out to an ally to ensure the client’s safety. A guide, on the other hand, is not likely to have access to some or any of these resources. Even if a name and address are available, triggering a police welfare check is an invasive option, which requires a substantial violation of privacy, and should only be invoked when necessary to prevent immediate harm.

How does a guide navigate this scenario responsibly and with the best interests of her user in mind? There is no easy answer that balances user safety with privacy and maintenance of the guide’s unique role. As demonstrated here, risk management cannot begin reactively at the level of the guide; decisions about the program’s model and the user selection and assessment strategies discussed in previous sections directly impact how risk is addressed. For example, if this user was accessing the program through a stepped care model, the guide could access support or coordinate care with his prescribing physician or primary care team. A more thorough assessment prior to starting the program could have alerted the guide to potential risk factors, and a safety agreement could have been developed at the outset of the IGSH treatment.

A final clinical risk issue to consider is how to manage treatment failure in the case of worsening symptoms or failure to improve. Psychologists have a responsibility to monitor response to an intervention on an ongoing basis and must adjust or terminate services they believe are not helping or are harming the client. They should provide referrals or help the user connect with alternative services, as appropriate (American Psychological Association, 2013). Although this can be a murky and subjective question for in-person therapists as well, here it is further complicated by the self-determined pacing of many IGSH programs; 4 weeks of participation in a program could look very different depending on whether a user was logging on daily or biweekly. The re-assessment of user progress may be done on a time-bound schedule or based on progress through the intervention. Where program access is based on a monthly subscription, it may benefit users to make IGSH programs time delimited as a matter of course, rather than allowing for indefinite participation over the course of months and years, while also allowing for the option to extend a program flexibly for a given user’s situation as needed.

5.1 |. Takeaways

Handling clinical risk in the context of an IGSH program poses unique challenges related to the format and structure of communication and the ability to access or not access crisis resources. Given that proponents of IGSH have emphasized accessibility of care as a guiding principle. An additional challenge is how to balance this value with potential harm posed by anonymous care or the reduction in other barriers to care, such as strict exclusion criteria or formal assessment. As this field grows, it is worth noting that a risk calculus may look different in the context of large-scale systems and the high-user volume that IGSH programs aspire to serve—an unlikely adverse event in a research study becomes a statistical inevitability when systems are serving hundreds of thousands or millions of users, meaning that programs and guides must be well-prepared and supported to address them.

6 |. RISKS, BENEFITS, AND INFORMED CONSENT

Informed consent is considered the primary means by which psychologists protect patients’ rights to privacy and choice of care that we are obliged to uphold (American Psychological Association, 2010; Fisher, 2017); it is a foundation of ethical practice whose importance cannot be overstated. Required components of informed consent include information about the nature and anticipated course of the intervention, fees, the involvement of third parties, and confidentiality, including rights and limitations. Services provided online, such as IGSH, are subject to additional guidelines related to associate unique risks and benefits.

Compared to offline treatment options, IGSH may offer significant advantages in terms of convenience and accessibility; compared to unguided self-help, it offers increased likelihood of treatment adherence and improved outcomes. However, it may also be uniquely vulnerable from a confidentiality and data security perspective. The APA Telepsychology Task Force identified patients’ understanding of increased risks to security and confidentiality, as well as steps being taken to protect confidential data, as being critical to informed consent (American Psychological Association, 2013). Part of the informed consent discussion should include education for users about steps they can take on their end to protect their own information safety. This may include setting a secure password that they do not share with others, logging out between sessions, and how to turn email or push notifications on and off. If users are not comfortable with the limitations presented by IGSH, they should be provided with access or referrals to an offline treatment option.

Service disruptions due to power failure, bugs, and Internet outages also warrant special consideration. Although these may be considered mere annoyances in some industries, when technology is used to provide emotional and psychological support, these unexpected disruptions can have a significant negative impact on users’ well-being. Consider, for example, users who regularly rely on their guides to help talk them out of doing something impulsive or dangerous, or who are suddenly in need of grounding techniques to ride out urges to self-injure. The informed consent process provides an opportunity to prepare users for the possibility of service disruptions and develop a plan to access secondary support services in such situations.

Informed consent should be conceptualized as a dynamic process that involves coming to an understanding with patients; the signed document (or its digital equivalent) should be viewed as documentation of this mutual understanding, rather than as the agreement itself (American Psychological Association, 2010). To that end, informed consent should be presented in a way that is accessible and understandable to the layperson. This traditionally involves discussion of terms at the outset of the first session, as well as a review of the relevant paperwork. However, it is not a one-time event, and the agreement should be revisited as necessary throughout the intervention. Examples of this ongoing process could include clarifying the boundaries of a guide’s role if a user appears unclear, or negotiating a safety plan in collaboration with the PCP that allows a patient to safely continue in the IGSH program.

Preserving the spirit of informed consent in the IGSH format presents new challenges and opportunities. To the extent possible, having live, in-person discussions with prospective users is the least risky, allowing for easy, inthe-moment responses to clarifications and questions. It also allows providers to more easily verify users’ understanding, or pick up on verbal or nonverbal cues indicating confusion. However, depending on the program model, this will not always be possible. Other options include offering an introductory phone call or providing a video or written introduction prior to beginning the intervention. These last two options should be crafted with care to use simple, clear language and avoid legal or technical jargon, and should include some pathway to seek additional information or clarification. One example of an innovative informed consent procedure for an online depression study combined custom YouTube videos with written material, and tested each user’s understanding via a quiz that addressed the voluntary nature of the study, and the fact that participants would be randomly assigned (Anguera, Jordan, Castaneda, Gazzaley, & Arean, 2016). Following initial consent, it fell to the guide to maintain ongoing consent throughout the intervention.

6.1 |. Takeaways

The requirement to obtain informed consent is not fulfilled by simply having IGSH users agree to a terms of service agreement online. This format requires creative solutions that help users understand what they are agreeing to, and empower them to make decisions about their treatment. Requirements regarding telehealth informed consent vary from state to state, and organizations engaged in interjurisdictional practice are responsible to be aware of laws in any state they serve, both generally and specifically, regarding providing online services. For a comprehensive account of laws regulating telehealth in all 50 states, see Epstein Becker Green (2016).

7 |. MOVING FORWARD

7.1 |. Resolving jurisdictional issues

One of the most exciting promises of IGSH technology is its potential to reach a large number of people at low cost to health care systems and consumers. However, jurisdictional issues discussed above are likely to present significant challenges to IGSH programs’ ability to scale at a national level in the USA. Clinicians and supervisors cannot realistically be expected to obtain licenses in fifty states, and maintaining a workforce of locally licensed providers in every state is prohibitive for all but the largest companies. This could impact access in states that are more rural and have less overall access to mental health care.

We posit that the current system is overly restrictive without significantly improving public safety and are concerned that the best interests of IGSH users will not always be well-served by the most obvious workaround—hiring only unlicensed guides such as health coaches or peers who are not subject to the same oversight. Legal solutions such as the Psychology Interjurisdictional Compact (PSYPACT), an interstate compact which would permit cross-state provision of telehealth services by licensed psychologists among member states (Association of State and Provincial Psychology Boards, n.d), are in the works, but implementation and widespread adoption are still years away. We applaud this effort and encourage professionals from other disciplines to prioritize similar legislation to allow for online practice across state lines.

7.2 |. Defining practice standards

Although IGSH falls broadly under the telehealth umbrella, practice guidelines for telehealth in the USA and abroad tend to focus more on “traditional” video-based telepsychology services. We are excited to be on the cutting edge of applying new technologies, but are also sometimes challenged to translate existing ethical guidelines to this format. Questions such as the appropriate credentials of guides and their supervisors; what levels of pathology can be safely and appropriately addressed using IGSH; and how to screen and monitor risk effectively throughout an intervention require additional empirical attention. The telehealth community would benefit if researchers agreed to share a higher level of detail in describing IGSH programs, such as specifics about guidance provided, supervisory structure, and risk management procedures in place, such that clinicians and technology innovators can learn from these and use data to develop best practices. We also call on professional organizations and state boards to broaden their understanding of telehealth to include IGSH and other emerging technologies and create or utilize existing task forces to examine these issues in greater depth. It is essential that organizations and clinicians have ethics resources that keep pace with technologies’ rapid growth and change.

7.3 |. Collaboration across disciplines

Finally, we believe that psychologists and other mental health professionals have a responsibility to collaborate effectively with professionals from other fields—engineers, product managers, marketers, content writers, designers, and others—who make IGSH programs work and disseminate them to the world. Our experience working on mental health technology has revealed cultural differences between these fields that have at times been a challenge to navigate. Anyone who has provided healthcare in a for-profit setting is familiar with the tension that can emerge between providing highest quality of care and making the best business decision. Similarly, psychologists working in technology settings will inevitably experience tension between offering a program that is clinically and ethically solid with the need for an agile model that allows for rapid iteration. When mental health professionals work in the technology field, it is their responsibility to ensure the IGSH programs they offer are rooted in science and executed in an ethical manner. To do this, they must develop a working understanding of the language, needs, and values of the tech industry. They must also be alert to areas in which their colleagues need education and information regarding the mental health field. On the flip side, it is imperative that health tech companies not only elicit critical feedback from psychologists (and representatives of the relevant ethical or regulatory bodies), but also meaningfully integrate them into the decision-making structures.

8 |. CONCLUSION

With technological innovation in the mental space, rapid growth has led to growing pains. Although IGSH has a strong clinical evidence base, we are still in the early stages of examining ethical considerations that arise delivering mental health services in this new format. This article aimed to provide a preliminary review of these issues. Taken together, we find that IGSH programs are empowered to provide safe and ethical care when they (a) coordinate care and risk management effectively with other systems; (b) provide guidance that is competent and well-defined; and (c) reach users that are appropriate for and well-educated about the services they are going to receive. In operationalizing these goals, providers must grapple with the tension between furthering the mission of improving access to care for all, and instituting robust practices to address less-common risks to IGSH users. We hope that this review will spark discussion, debate, and further innovation as we work toward developing an industry standard for IGSH intervention.