All Articles
Insight of the week
Research

Can digital interventions replicate human communication?

Software programs like CIAS translate techniques from Motivational Interviewing into digital experiences

HTD Health

• 9 min read

Digitally-enabled care has taken center stage over the last two years as providers and patients alike have adapted to omnichannel healthcare delivery. The COVID-19 pandemic forced a rapid shift to remote care delivery, propelling existing innovation in virtual care services. While industry experts agree that there is an important place for human-to-human connection in healthcare, we have also seen a growing opportunity for digital tools and platforms to expand access to certain types of care. 

Motivational Interviewing (MI) has the unique ability to encourage positive behavior change among an incredibly broad population of patients. In the following article we will look at how motivational interviewing is being adapted to digital channels and what that means for many individuals who may not otherwise seek medical care for unhealthy or dangerous behaviors.

Sign up for the HTD Insights Newsletter

Healthtech research, news, and industry-defining trends delivered to your inbox monthly.

We respect your privacy. This email address will only be used to send HTD’s Insights newsletter. Click here to review our full privacy policy.

Thank you for subscribing!

Introducing Motivational Interviewing

Motivational Interviewing was initially introduced in 1983 by clinical psychologists William Miller and Stephen Rollnick as a way to increase patients’ motivation to change their behavior. The practice encourages patients to reflect on the factors that influence their own decision making and take a more active role in evaluating their behaviors. Miller and Rollnick describe the “spirit” of MI, which is based on three key elements: 

  1. Therapists collaborating with their patients
  2. Enhancing intrinsic motivations for behavior change in a patient
  3. Accepting the patient’s right to autonomy and ability to make their own changes

What is unique about motivational interviewing is that it encourages people to think through their own health behaviors and decide whether they would like to make a change. This means that the technique has the potential to reach patients beyond just those who are treatment-seeking. Dr. Steve Ondersma, a clinical psychologist and professor at Michigan State University explains: “We’re talking with folks about whether or not they might want to make a change rather than specifically how to make a change. And instead of waiting for people to come to us, we’re going out and proactively starting these conversations.” 

Therapists encourage ‘change talk’—nudging a patient toward action—by demonstrating empathy, highlighting discrepancies between the patient’s desired states and current behavior, and promoting self-efficacy. Since its inception nearly 40 years ago, the principles and methodologies of MI have been implemented and tested in numerous settings, establishing it as an effective, evidence-based practice to elicit and strengthen motivation for behavioral change. In an analysis of 72 randomized clinical trials that reviewed the effectiveness of MI in encouraging behavioral change, Rubak et al discovered that MI had a significant and clinically relevant effect on modifying behaviors in approximately 75% of the studies.

The digital opportunity for motivational interviewing

In its original form, motivational Interviewing was conducted through one-on-one, in person discussion. The time and resource intensiveness of the method makes it difficult to scale MI to broader populations: “These relatively brief, proactive motivational interviewing-based approaches normally take place in a primary care setting,” explains Dr. Ondersma. “But that can be really difficult to implement. You have to specially train nurses and doctors and find a way to carve out time from an already overbooked, busy clinic flow.” Instead, researchers are increasingly turning to digital interventions that adapt MI in order to reach a larger audience. 

According to research by Ondersma and his colleagues, digital interventions or e-interventions are often equal to traditional in person interventions regarding treatment attendance, session completion, retention, and patient outcomes. Further, e-interventions are deemed acceptable to users with regards to patient satisfaction and ‘therapeutic alliance,’ broadly defined as the overall relationship between therapist and patient during the process of therapy. Similarly, technology-based delivery of screening, brief intervention, and referral to treatment (e-SBIRT)—a practice based on MI—has shown potential in several studies. In a trial by Schwartz and colleagues, it was revealed that e-SBIRT was no different from provider-delivered SBIRT in primary care when using a global drug use and consequences outcome, and was significantly better for marijuana and cocaine use outcomes. The e-interventions were also deemed ‘highly acceptable’ by patients. 

CIAS: An open-source tool for building digital interventions

In his research, Dr. Ondersma develops, validates and implements electronic healthcare interventions to identify and address non-treatment seeking individuals with behaviors such as unhealthy substance use. Dr. Ondersma has developed CIAS (Computerized Intervention Authoring System), a digital platform and research tool equipped with motivational techniques. This open-source, low-code software acts as a “blank slate” for researchers to create interventions using any theoretical model including motivational interviewing, cognitive behavioral therapy, etc. Academic researchers can easily build digital interventions using a variety of pre-built features and deploy them to participants via a web application.

CIAS can replicate many motivational techniques in a digital setting: Survey questions ask patients to describe their behavior and their feelings about it. Then, using a non-judgmental approach, additional screens help them identify the possible benefits of change and offer the opportunity to set a behavior change goal. At the end of a session, if an individual has self-identified a particular behavior they wish to change, they are offered additional resources.

CIAS is unique from other digital health intervention technologies and survey-building tools in the space in that it: 

  • Allows content experts to develop interactive mobile web apps without coding
  • Features an emotive narrator who can provide natural-language reflections
  • Presents highly personalized content of any kind whether text, data visualization, image, or video

Adaptability of the CIAS platform

Unlike other applications that focus on a singular health outcome, CIAS can be adapted to target many different behaviors, from smoking and illicit drug use to exercise and healthy eating. The technology also enables researchers to screen for multiple problems in one session, by embedding specific behavioral questions such as alcohol or drug use within a multitude of other health-related questions. 

Reaching broader populations

CIAS employs MI as an evidence-based approach to reach patients who, despite needing treatment, neither receive it nor seek it out. It encourages self-change among those whose behavior does not meet criteria for serious concern, but presents a health risk nonetheless. This audience is much larger than just active treatment-seeking patients, making CIAS a powerful tool for population health. 

For example, 21.7 million people need substance use treatment each year in the US. However, only 7.9% of those with harmful alcohol consumption habits actually receive specialized substance abuse treatment services and among those who do not, many are not aware that their behavior is harmful. Similar findings from Schumacher and colleagues estimate that the annual prevalence of alcohol, drug, or combined alcohol-drug use disorders is 9%. However, only around 10% of those who meet the criteria for these disorders receive treatment of any kind. Through CIAS, Dr. Ondersma and other researchers can implement proactive motivational approaches across non-treatment seeking, at-risk groups and reach far greater proportions of the population that need support than could ever be reached in person. 

Instilling MI best practices

Aside from the high reach that technology enables, software like CIAS can also bring MI approaches into busy healthcare settings where they might otherwise be difficult to implement. Person-delivered motivational interventions take considerable time, money, and logistics to integrate into ongoing medical practice. 

The psychology of designing computer-delivered interventions

While the potential for a tool like CIAS is vast, developing digital interventions that can replicate some degree of in-person delivery is challenging. Dr. Ondresma explains: “While it’s absolutely true that computers can’t do motivational interviewing and replicate that exact interpersonal experience, it’s also true that computers can engage people on a level that is important. But you must be intentional about the way that you integrate social psychology into the technology.” This perspective mirrors HTD Health’s approach to thoughtful digital transformation: Technology alone cannot solve healthcare, but when approached with users’ needs in mind, it can exponentially increase the reach and impact of people. 

While designing technology that supports MI techniques, there are several fundamental psychological principles that designers and developers must consider to ensure the same levels of participant engagement and effectiveness as face-to-face MI approaches. An entire division of research that investigates human interaction with computers demonstrates that as soon as you give a program a human feature such as a voice, personality, character, narrator, or face, users automatically respond to it in a similar way to how they interact with a living, breathing person. Although no one is fooled that they’re working with a computer, unconsciously (in terms of engagement) they respond to it as if it is human. This phenomenon is known as anthropomorphism. 

Clifford Nass, Stanford social psychologist and computer scientist, dedicated his career to testing the body of scientifically validated principles of social psychology found in human-to-human communication to determine whether they can be replicated in interactions between a person and a computer. Time and time again, these social principles were found to stand. 

For example, when people interact with a computer that flatters them, they prefer that computer and want to interact with it more. While computers can’t replicate the exact interpersonal experience of face-to-face interaction, teams can design software to engage people on a significant level. The following elements help patients connect to digital interventions in ways that mimic in-person interaction:

  • Narrator: Building a lifelike presence—even a simple cartoon character—into the software with whom participants can interact increases emotional connection and engagement. Studies show that the mere presence of an embodied human-like agent (simulated characters that represent human-like qualities) have a positive effect on user engagement. In one study, participants were given feedback about their performance in two ways: once with just words and the other with words next to a picture of a yellow robot. Participants rated yellow robot feedback as “more helpful” and stuck with the task longer.
  • Voice: Including voice technology into the software further reinforces the emotional connection. Much of Dr. Nass’ early research focused on software that involved nothing more than a simple voice that talked to participants. This demonstrated that again, people behave toward voice-based technology in similar ways as they do toward interactions with other people. 
  • Synchronous interactivity: Software should appear to respond in real-time to the participant’s activity like one would expect in an in-person conversation. For example, CIAS limits a participant’s inputs to multiple-choice, so researchers can respond based on their answer in a reactive, emotive way that resembles the turn-taking in a person-to-person conversation. After one question, rather than jumping straight to another question, there’s a character that talks out loud and says something like “From your perspective, it would be tough to quit smoking because it’s how you deal with stress, and all of your friends smoke too. You also feel that you’ve tried to quit multiple times before and it didn’t work, but you worry about how much money you spend on cigarettes and the effect on your health. Have I understood that correctly?” This response is understanding; it’s alive and talking to the participant in a human way.

New applications for CIAS

HTD Health has been working with Dr. Ondersma to design and develop a new version of CIAS with improved UX and added features. The CIAS 3.0 project is supported by the NIH and is intended as an open resource for scientific research. With the new version launched, Dr. Ondersma has been thinking about new applications where CIAS could be leveraged: “In a healthcare setting such as an outpatient clinic, or in my case, a prenatal care clinic, we’re giving patients an iPad and asking them to run through the program. In that setting, 90% of the patients with access to the program are using and completing it. It’s really exciting that we’re getting those sorts of numbers proactively, without needing a research assistant in the clinic.” 

In Dr. Ondersma’s ideal world, the technology would be integrated into national healthcare systems to reach as many people as possible in the most effective way. However, this would be no easy task: Technologies such as electronic health records took billions of dollars and many years for adoption in clinics across the US. Despite that, if US laws change to allow healthcare systems to bill for screening and brief interventions for more than just substance use, many more clinics could adopt the technology. 

“We’re seeing rapid adoption of patient-facing medical record apps like Epic’s MyChart and others,” adds Dr. Ondersma. “The more adoption of these apps grows, the better our opportunity will be to include CIAS in such a platform and get interventions in front of as many people as possible. Other potential application settings could include integration with employee wellness programs, whereby employers can provide incentives for employees to engage with this technology. We could also target smaller, but very high-risk groups, such as incarcerated people.”

Dr. Ondersma stresses that digital interventions such as those created with CIAS will not replace in-person care, but will instead provide a new layer of patient support and resources that might otherwise not be possible in the clinic setting. “It’s not a replacement for in-person care, it’s a replacement for the blank space where these conversations currently aren’t happening. When people have a healthcare encounter, too often nobody asks about these things. And if they ask, they probably don’t address these behaviors because they can’t—they’re too busy, and they’re overwhelmed.” Moreover, as the technology is used more widely, he hopes for large, collaborative trials that will help identify the best ways to help people and adapt the technology accordingly. 

Given the capacity of digital tools to reach vastly greater numbers of people with health-related interventions, MI techniques delivered through CIAS could have monumental positive impacts on public health and wellbeing. By empowering patients to make their own decisions about health behavior change, motivational interviewing can bring about greater levels of engagement and commitment to rethink unhealthy behaviors such as smoking, alcohol, and drug use. As CIAS and similar virtual care platforms are introduced and adopted, it will be a fascinating area of medical technology to watch. 

HTD Health

HTD Health is a digital services group working with the healthcare and wellness industries. HTD's experienced team works with clients to plan, design, and build custom virtual care software.