How to influence patient health behaviors: Virtual care opportunities

In recent years, there has been an increased focus on changing behaviors to align more closely with healthy habits. There is no shortage of apps on the market that tout their ability to help consumers change and sustain a behavior. Many of the strategies for promoting behavior change draw from behavioral theory and target motivation and engagement to sustain a habit, thus positioning consumers at the center of the process. Despite this, many interventions lack a well-supported scientific, theoretical basis for their techniques—a host of moving pieces rooted in the inherent diversity of the target population make it difficult to isolate a specific strategy or set of strategies that can be generalized to apply to a whole population or a particular behavior.

However, with the rise of digital technology to supplement existing practices and bridge gaps of in-person healthcare, the future of behavior change-related digital interventions will likely see immense growth and transformation, leading to gradual changes in population health in the  United States.

This article will look at prevalent behavioral theories, the relevant techniques they inform, and the emerging space of digital interventions for behavior change.

There are six theories and models of behavior change commonly referenced in the healthcare literature—the health belief model, the social cognitive theory, the social norms theory, the theory of planned behavior, the self-determination theory, and the transtheoretical model. While these theories originated in psychology, they are hardly contained within just this discipline; applications to education, business, and healthcare are becoming increasingly prevalent. In healthcare, much of the behavior change research is specific to a particular field, such as this review of health behavior theories as they relate to efficacy in promoting medication adherence for tuberculosis and HIV/AIDS, this review on clinical interventions in aging, and this review of behavior change techniques for diet and physical activity in type 2 diabetes.

Under the health belief model, a person’s motivations for action depend on the perceived benefits and barriers related to the behavior. There are six constructs:

• Perceived susceptibility: This refers to a person’s belief of the likelihood that they will contract an illness or disease.
• Perceived severity: This refers to a person’s belief of the extent of the seriousness after contracting an illness or disease.
• Perceived benefits: This refers to a person’s belief of the effectiveness of taking preventive actions to prevent or reduce the likelihood of contracting a disease.
• Perceived barriers: This refers to a person’s perceptions on the obstacles they may face in making a health behavior change.
• Cue to action: This refers to the catalyst that stimulates the person into action.
• Self-efficacy: This refers to a person’s confidence in their own ability to perform a behavior or action.

The social cognitive theory posits that people learn through observing the behaviors and outcomes of others. According to this theory, people are constantly influenced by interconnected personal, behavioral, and environmental factors. This concept is known as reciprocal determinism. As its name suggests, each determinant impacts the other, effectively forming a network where one influences and is influenced by the others.

The other foundational constructs of the social cognitive theory are:

• Behavioral capability: The ability to carry out a behavior, which may be contingent on adequate education and training to know what to do, as well as the means to know how to do it.
• Observational learning: The modeling of a behavior, in that a person is able to reproduce behaviors that they observe.
• Reinforcements: Closely relates to how the outcomes of reciprocal determinism—internal attitudes and environmental factors, both positive and negative—promote a behavior.
• Expectations: The anticipated outcomes that are subjectively determined by the individual and may be influenced by observation and past experiences.
• Self-efficacy: Refers to the individual’s own confidence and self-perception that they can carry out a behavior, and is influenced by both internal and external factors.

As each individual is exposed to external and internal influences, they must self-regulate to avoid simply conforming to the social and environmental factors at play. Self-regulation allows people to control how they react, both emotionally and behaviorally, to the environment and other external stimuli. This provides the basis for purposeful action. People are empowered, through proactive and self-regulating actions, to learn and shape their behavior. This includes self-monitoring of behavior, judgment of behavior in relation to the circumstances, and self-reaction to those observations.

The influence of a person’s social environment continues to be a common thread in the social norms theory, which focuses on correcting the gap between perceived norms and actual norms. People tend to overestimate the level of risky behaviors and underestimate the level of healthy behaviors their peers engage in, leading them to alter their own behaviors to fit in with the perceived norm. This theory has been used to explain behaviors such as smoking and drinking in adolescents, who are at a particularly impressionable age.

The theory of planned behavior builds upon existing concepts from developed theories—it combines the elements of personal attitudes, subjective norms, and perceived control. To break down the components of this theory further:

• Personal attitudes include both behavioral beliefs and outcome evaluations. Behavioral beliefs are beliefs about the likely consequences of that behavior, and outcome evaluations refer to assessment of those consequences.
• Subjective norms include normative beliefs and motivation to comply. Normative beliefs are beliefs about the standard expectations of others, and motivation to comply refers to the outcome of an assessment determining how important it is to follow what others expect.
• Perceived control includes how much a person has control over the behavior and how confident they feel in performing it

Together, these factors contribute to an individual’s intentions, which predate their behavior. This theory has been used to predict many health-related behaviors, including smoking, drinking, and exercise.

Self-determination theory focuses on the motivations behind behavior change. According to self-determination theory, not only do social factors influence motivation, but there are different types of motivation which are more or less successful at driving long-term behavior change. As described in this theory, there are a range of motivation types that all fall on a spectrum of external or extrinsic motivation. This means that the motivation comes from outside the individual—from expectations, rewards, etc.

External regulation is the least autonomous, in which behaviors are regulated by compliance, conformity, and external rewards and punishments. Behavior change influenced by external regulation is not likely to last long-term because as circumstances change, there is little driving an individual to keep up that given behavior. Introjected regulation is still somewhat externally motivated, but is also associated with internal rewards and punishments. At this stage, people are often motivated by feelings of guilt or by self-inflicted pressure to prove self-worth. In identified regulation, motivation is more autonomously driven, and it involves conscious values that are important to the individual. Integrated regulation is the most autonomous type of extrinsic motivation, in which the behavior is integrated with personal beliefs and values. Movement from external to integrated regulatory behavior and towards intrinsic motivation is known as internalization, and underlying this process are the three psychological needs of autonomy, competence, and relatedness.

• Autonomy refers to feeling like one has a choice, rather than being subjugated.
• Competence refers to a person’s capability to achieve an outcome.
• Relatedness refers to connections with others and a sense of belonging.

Internalization is optimized when social conditions favor the fulfillment of these needs. This is important because internalized motivations are generally more effective at promoting behavior change because these factors are the ones that are controlled by the individual, whereas extrinsically motivated factors lack that degree of personal control.

The transtheoretical model depicts the stages of the behavior change cycle: precontemplation, contemplation, preparation, action, maintenance, and termination. This model outlines a framework for which various theories and techniques fit into.

• Precontemplation: At this stage, people are not ready to alter their behavior, and may not even be aware that behavior is problematic.
• Contemplation: People are aware of problematic behavior and begin to look at the pros and cons of changing their behavior. Even though these considerations are being made, people are not yet ready to take immediate action.
• Preparation: At this stage, people are ready to take immediate action, and they begin to make incremental changes.
• Action: During this stage, people have been and are continuing to take action that leads to behavior change.
• Maintenance: At this point, people have sustained the behavior for at least six months, and they intend to continue maintaining it.
• Termination: By this stage, people are confident that they will not return to old behavior.

People are at different stages of readiness for change at each stage of this cycle, opening up opportunities to create interventions catered to the stage of change. For instance, schemes in the precontemplation and contemplation stages may include educational interventions aimed to improve understanding of the negative consequences or risks of unhealthy behaviors, while strategies for the preparation and action stages may be geared more toward goal setting and curating an action plan. Interventions in the maintenance stage may involve periodically following up with the individual and providing prolonged support and regular re-evaluation and assurance.

Although these major behavior change theories were developed in tandem with one another, the interplay between personal and environmental factors is a common thread, which underscores its significance in behavior change.

Despite the fact that behavior change theories have been established and circulated for years within the fields of psychology, sociology, economics, and more, there remains something of a “black box” of behavior change—what exactly are the mechanisms that promote and sustain behavior change?

Unsurprisingly, there is not one technique or even one set of techniques that serves as a silver bullet for this layered and nuanced question. Not all interventions work for everyone all the time. Still, from a macro-level review of the literature on digitized interventions, three techniques emerge as common strategies to optimize positive behavior change:

• Goal setting
• Self-monitoring
• Social support

Goal setting requires a person to have control over the outcomes that they are working toward and an understanding that they are capable of achieving that goal. This plays into the importance of self-efficacy from the social cognitive theory, as well as the internalization of motivation from the self-determination theory. Erin Commons, a registered dietitian and the Vice President of Care Management at Vivante Health, emphasizes this point: “It’s important to evoke the members’ own motivation and commitment to change rather than telling them what they should change.” Setting SMART goals—ones that are Specific, Measurable, Attainable, Relevant, and Time-related—motivates efforts toward concrete endpoints by identifying actionable steps. Commons underscores the reasons for a behavior change as part of the relevance aspect of a SMART goal. Vivante’s care team helps members understand not just a big picture desired change (e.g. lose weight or improve symptoms) but the deeper goals that underpin them—for instance, having more time to spend with family or not missing out on important life events. Once the root motivations are understood, Vivante’s care team can more empathetically and effectively work with members toward their goals.

At Monument, a personalized online alcohol treatment program, physicians and therapists similarly work with patients to understand their motivations for changing their relationship with alcohol and create a personalized plan to achieve their goals. Connor Walsh, Monument’s Director of Operations, explains: “Members’ goals extend far beyond achieving sobriety or moderation. They want to take back control of their relationships, finances, mental and physical health, and beyond. Personalized, evidence-based therapy can help them understand how alcohol affects all dimensions of their wellness, and take tangible steps towards reducing the role alcohol plays in each one.”

Once members have a deeper awareness of what influences their relationship with alcohol, and how their drinking affects their overall wellness, Monument therapists and physicians provide guidance and accountability for cutting back. “We meet people where they are,” says Walsh, “not everyone is aiming for complete abstinence. Others might begin tapering back with an ultimate goal of sobriety. Our relationship with alcohol is deeply personal, and treatment should be, too.”

Self-monitoring is often used in tandem with goal setting. The prevalence of this technique in behavioral health apps stems from the pillar of self-regulation described in the social cognitive theory. It is a method to continuously track progress and keep the user consistently engaged with their goals. Self-monitoring also pushes people to bear responsibility for their behaviors, rather than attribute their health, for instance, to external influences such as genetics. With the feature to track progress over time, it also fosters a sense of competition with oneself, which encourages people to strive for improvement.

At Monument, members are encouraged to engage in drink tracking to practice accountability. Drink tracking encourages self-reflection and allows members and their Care Team to understand progress over time. Tools like calculating money saved by cutting back on drinking also provide financial motivation and rewards.

Social cognitive theory strongly ties into the use of social support as a technique through which to promote behavior change. The theory claims that one way people learn is through social interactions with their peers. These interactions influence reinforcements and expectations for behavior and its outcome. The social norms theory may also be at play here. Membership in a group of people that is working toward a similar goal can help individuals understand where they fit in with their peers, reducing the possibility of misperceiving what is normal. This is further supported by the psychological need for relatedness, proposed in the self-determination theory.

Community is key to Monument’s treatment approach. Therapist-moderated support groups help patients work through challenges together while holding one another accountable for their progress. Monument also offers an anonymous (optionally) online discussion forum where members can post updates, questions, or challenges and receive positive support or feedback from others in the program.

Like with alcohol treatment, conditions connected to digestive health can come with a social stigma. At Vivante, for instance, they host webinars that share personal stories to bring individuals together. However, their platform remains flexible, recognizing that not all members are comfortable sharing experiences in a face-to-face video call. They encourage connection in any way that engages the member, including phone calls, in-app chat messages, and the aforementioned 24/7 staffed nurse call line. As Commons explains, “Building a strong rapport with members is important. I want them to know that not only are we going to help them with their behavior change, but we also really care about them as a person.”

All three techniques are often used synchronously, and it is likely that their overall effectiveness in promoting behavior change is synergistic in nature. And while the diversity of patient populations and social factors bring interesting nuance to behavior change techniques, it can also make it difficult to evaluate their effectiveness. There are few clinical studies that concretely point to the success of a certain technique or set of techniques to address one behavior within virtual care.

Furthermore, even within one behavior, there are multiple potential populations that may require different approaches. For instance, encouraging smoking cessation in adults may look different from encouraging it in teenagers. An adult smoker with a family may be more internally motivated to stop smoking if smoking’s negative health impacts on others were emphasized, effectively reframing it from an individual health concern to a relational one. A teenage smoker, however, is unlikely to be similarly motivated; instead, correcting misperceptions, as advanced by the social norms theory, may be more effective. Helping the teen smoker understand how few peers are actually participating in the unhealthy behavior may help correct their perceptions and lead them to alter their behavior accordingly.

The digital health ecosystem has grown rapidly in recent years, with the COVID-19 pandemic accelerating the adoption of digital medicine. Accompanying this shift toward the digital are novel design considerations and significant advantages, both of which are highlighted next as they relate to driving behavior change.

Building an app begins before product development. It is important to set a clear vision, outline goals, and understand the target population before starting the design process. Depending on target population demographics, design considerations to optimize user engagement and experience can vary. Even within a space that has been established and researched, there may be missing pockets of information, simply because each situation is different. This makes user research a valuable tool to understand and address the needs of a specific patient population. Without user research, the design process is based on assumptions that may produce a subpar experience and require more product updates later on.

Amy Bucher, author of Engaged: Designing for Behavior Change, explains the steps involved with building a successful intervention: “We work backwards from the desired outcomes and we figure out what behavioral changes we need to see in our population in order to get to those outcomes. And then even before that, what is the intervention we need to deliver to support those behavior changes? Working backwards is a way to ensure that whatever we deliver as an intervention speaks directly to those behaviors.”

Additionally, a host of other considerations, including the user interface, marketing, and deployment strategies, all need to be examined. This includes how the intervention is delivered, how people receive it, and how people make use of it. With a switch from in-person interventions to mobile platforms, notifications, alerts, and other distractions that pop up may detract from the optimal level of user engagement, and as such, must be accounted for. Although consistency is key to lasting habit formation, apps that strictly follow this guiding principle and push users to regularly track their progress or do daily check-ins on the app risk losing users to quitting prematurely. Once a day is missed, it can be hard to get back on track, adding stress to an already unfamiliar situation. Instead, incorporating the intervention into existing habits reduces the likelihood that it is perceived as a burden.

Despite the lack of clinical data to support the efficacy of digital platforms, virtual interventions come with several advantages:

• Streamline goal setting and allow users to frequently revisit goals
• Record and present tracked progress in appealing visuals
• Promote engagement and accountability through building community

The rising ubiquity of internet access and mobile phone usage has paved the way for theory-based intervention research to use existing apps to expedite data collection—data from apps that utilize a continuous progress tracking function can easily be exported to provide longitudinal data that tracks long-term changes. This data can also be exported to healthcare professionals to further integrate health apps into primary care.

The versatility of digital platforms to harness video, audio, and text capabilities to increase user engagement also holds an advantage over traditional in-person interventions. Online social communities are also more accessible—they lower the barrier to entry by removing the transportation hurdle and lessening the burden and commitment to attending. As posited in the social cognitive theory, people learn by observing the environment and their peers, so the ease with which an online community can be formed is conducive to positive behavior change.

Many digitized interventions combine multiple techniques and aim to provide a holistic approach to behavior change.

Calibrate brings an interdisciplinary approach to a weight loss program—it combines behavioral interventions with biological one. Users are prescribed metabolic medication, assigned to a doctor and accountability team, and tasked with self-monitoring their weight, sleep, food intake, and physical activity. It is a year-long program, and over this time period, users are prompted to make incremental changes in their lifestyle to improve their metabolic health.

Monument is an online alcohol treatment program that operates under a model of providing specialized treatment. In each of the three paid plans, users are connected to a physician that can prescribe two FDA-approved medications—disulfiram and naltrexone—to aid in changing drinking habits, and in the biweekly and weekly Total Care plans, users have access to a therapist as well. Users can also opt for a free Community Membership that provides access to group support. The therapist-moderated support groups are hour-long virtual sessions that cover a broad range of topics relating to changing drinking habits, and there are sessions tailored to specific groups, such as women, BIPOC, and trans or gender-nonconforming individuals.

GIThrive from Vivante Health is a digestive digital health program that combines a clinically rigorous, evidence-based platform with guidance from a dedicated care team to drive a personalized approach for improving gut health. Commons identifies key aspects to their approach: understanding motivations behind behavior change is at the crux of sustaining that change, establishing a network of support provides assurance and accountability, and foundational to both of these components is a strong relationship with the member.

Omada Health offers digital care programs for type 2 diabetes, hypertension, prevention, physical therapy, and behavioral health. Following its tagline of “bringing human connection to digital care,” it uses evidence-based interventions enhanced by user design for seamless integration into everyday life and augmented by a personal care team for feedback, support, and accountability. Its website features case studies, white papers, and peer-reviewed publications that provide evidence for its interventions.

Noom is a fitness and weight loss app that sustains user weight loss outcomes by encouraging behavior change through regular self-tracking, educational content, and motivational quizzes. A strong scientific basis is at its core—a sizable portion of its website is dedicated to accessible research, including peer-reviewed articles and internally produced insights and analyses.

Tempest is a recovery program that offers a Core Membership and an additional Intensive program, both of which plug members into communities where they can share their experiences with others in recovery. The memberships differ in their level of engagement; videos, workbooks, and recordings are available for the full year for the Core Membership for users to utilize at their own pace, whereas membership in the Intensive track involves weekly lessons and meetings. While the Core Membership teaches practical skills, such as managing anxiety, setting boundaries, and navigating holidays, the Intensive add-on touts a clinically-proven program to becoming sober and goes a step further and targets the member’s root cause for drinking.

Providing holistic care is a common theme across these apps, and many companies integrate community support as well as expertise from medical professionals into the behavioral strategies of the intervention. And while these companies employ strategies rooted in science and research, that science and research remain largely unpublicized, further sequestering the black box of behavior change from the public eye. While it makes sense for companies to protect their proprietary information, this then pokes at the balance between leveraging company-specific, evidence-based research as a marketing strategy and keeping that information private. It also raises larger questions of the significance and tactics of marketing toward consumers, such as how much value consumers place in a research-based intervention and how much a company can reveal about their research without revealing too much.

With the surging interest in the digital health landscape, the stage is set for digitized behavior change solutions to explore novel ideas as well as strengthen the currently tenuous connections between theory and practice.

One goldmine of opportunity is precision medicine’s analogous arm in behavior change. These types of solutions recognize the heterogeneity of the population, even among people with the same condition, and move away from a one-size-fits-all approach. For instance, many research groups are exploring how genetic and environmental influences play a role in human health, giving rise to fields such as pharmacogenomics. In the behavior change space, an emerging field of interest is tailoring interventions to individuals through personality matching, the idea being that people with similar personality dynamics may benefit from similar types of interventions. Looking at the big five personality traits—openness, conscientiousness, extroversion, agreeableness, and neuroticism—is a starting point for this approach. Experts are also exploring other metrics, such as morals and personal values, to measure human traits that carry a currently indeterminate indication toward intervention efficacy.

Another source of opportunity is “recycling” behavioral change techniques in the grander scheme of health interventions. For instance, Omada Health’s initiative stems from an original pilot program that was then adapted to fit other health conditions. This suggests that a core strategy can be effective in targeting multiple conditions. This approach to reach a broader population does not contradict the push for personalized care; in fact, it emphasizes it. Having dedicated specialists that personalize treatment for the member is a common denominator across the programs offered by Omada Health, indicating the significance of an individualized care plan. The question then is not whether personalized care and broad-reaching care are in opposition with one another, but rather what makes a condition appropriate for adaptation of a core strategy.

Additionally, companies must also evaluate the feasibility of these human-to-human interactions as these approaches gain traction and companies scale up to meet demand. Technologies such as chatbots enhanced by artificial intelligence are used to support behavior change interventions, but embedded in this intertwining of technology and design is the challenge of finding a balance between the human and the digital—how do companies use technology to augment user experience without sacrificing the elements of personalized, human care?

Looking more generally at behavioral science in the workplace, digital tools to improve health outcomes sit at the intersection of understanding human nature and developing cutting-edge technology. Bucher expressed excitement at this opportunity niche, saying that more and more, behavioral scientists in digital tech companies are pioneering positions that didn’t exist before. Social scientists offer expertise that can supplement the hard sciences and provide a human-centered lens for approaching a problem. For instance, in artificial intelligence, the algorithm is only as good as the data, and history has shown that unrepresentative datasets are a glaring issue when the model is imposed on an overlooked population.

In the behavior change workspace, biases in interventions can be mitigated by incorporating a multidisciplinary approach because experts from different fields naturally look at the problem in different ways, thereby informing the work from multiple standpoints and reducing the risk of introducing biases. The Behavior Change for Good Initiative is a collaboration between experts from the social sciences, medicine, computer science, and neuroscience to tackle the challenge of sustaining behavior change. Similarly, the Human Behavior Change Project brings together behavioral scientists, computer scientists, and systems architects to build a “Knowledge System” that will ultimately aid in understanding the underpinnings of behavior change interventions. This multidisciplinary collaboration between the hard and soft sciences is an emerging trend, and if its predecessor of cross-discipline teamwork in healthcare settings is any indication, it will help improve outcomes both for consumers and in the workplace.