Clinical Researcher—April 2020 (Volume 34, Issue 4)

SPECIAL FEATURE

Mindy Gruba, MPH

 

Changing behavior—to some, it’s “second nature.” You just do it. As you grow up, you learn to pay your bills on time, you start calling your mom every Sunday, and you recognize the value of creating to-do lists. However, for some groups, or for more complex changes, the idea of changing behavior provides challenges each step of the way.

As a society, we sometimes attribute lack of follow-through to just forgetting. While that’s absolutely valid, people don’t just do or not do things because they forget, or because they remember.{1,2} In fact, people follow through on things because it’s easy (online shopping), there’s risk if they don’t (paying bills so they can continue to use their phone), their friends are doing it, or simply—they just enjoy it.

In the context of healthy behaviors, there are many tactics that health professionals employ to motivate their patients. These may include promoting tiny habits, such as “prescribing” a single, simple exercise, or explaining how medication can improve their symptoms. Patients may be especially driven when they start to experience the outcomes of being compliant, such as a reduction in disease symptoms.

With all that said, the ecosystem of what influences behaviors is highly complex, and can’t always be attributed to a single thing. In the context of clinical research, taking a “health behavior” approach to help the person lends to this complexity. The intent of the trial isn’t necessarily to make a person better, it’s to answer questions. What makes this even more tricky is that patients in trials may not be getting better, either because they’re in a control group or because there’s no proof of effectiveness of the intervention.

How are Behavioral Tactics Currently Used?

With the rise in behavioral economics, we’re now seeing industries begin to use and implement psychological insights to influence behavior through minute levels of persuasion, instead of coercion.{3} It’s clear that this has improved desired outcomes for those who are implementing these tactics.

For example, there are some software services that default to a more expensive, yet unnecessary option upon checkout. The preselected offering isn’t something you would have proactively chosen, but when it’s the default, it feels like the standard. This may lead you to exhibit loss aversion. You now perceive your originally intended purchase as inferior, and a “downgrade” to be avoided. This revenue-generating tactic still provides choice, but nudges consumers to paying more than planned.

Conversely, we’ve also been seeing these tactics applied for good, such as default opt-in for organ donations and retirement saving plans, thus saving lives and building financial security.

Constructs of Behavioral Change Models

A behavioral change model relies on constructs to drive a framework that helps to understand the psychology of why people do the things they do. This can be used to drive strategies on how to influence those behaviors in a more desired direction, similar to the tactics previously described.{4}

While there are many behavioral models, most share the same or similar constructs. Table 1 lists a few of the key ones, and how to independently incorporate them into your patient engagement plan.

Table 1: Key Constructs of Behavior Change Models

Construct	What Patients Might be Thinking	Strategies
Perceived Risk	What is the likelihood something bad will happen? If something bad happens, how extreme will it be?	Inform patient of risks, focusing on susceptibility and severity of the condition or behavior (fear appeal) to convey importance of diligence, but emphasize methods for prevention and treatment to overcome risk to reinforce perceived efficacy.{5}
Knowledge	Do I understand what is being asked of me and why it’s being asked?	Provide basic information about a medical condition that might include how the disease develops, its expected course, and how specific strategies can help manage it. Apply this same approach to study requirements, such as explaining why{6} patients need to take all their medication at the specified time.
Skills/ Ability	Can I stay organized and do I have the tools I need to be effective?	Provide patients with intuitive{3} take-home instructions and tools, such as easy-to-use pill boxes,{1} visit schedules, and preparation guides, along with reminders on their own phone. Help patients build good habits early in the study by anchoring to known habits that already exist. For example, if patients need to take a pill once a day, you could instruct them to take the pill just before brushing their teeth.
Perceived Self-Efficacy	Am I confident in my ability to reduce the risk or attain the benefit?	Try the “foot-in-the-door” technique. First make a small, attainable request to build efficacy.{7} For example, have the patient complete a training diary at the site or review the technology with them so they will feel more confident about using it at home or on their own. Technology training is particularly valuable with older populations.{8}
Response-Efficacy	Will the outcome reduce the risk or increase the benefit?	Provide concrete examples of how risk could be averted.{5} In some trials, for example, a patient might be instructed to avoid certain medications. While it’s important to make them well aware of the risk, it should also be made clear that avoiding the drug can reduce the chance of a bad reaction. Giving specific examples, like brand names, makes it more concrete for the patient.
Subjective Norms	What are others doing? What do they think I should do?	Convey basic characteristics about others in the trial, such as how many are participating and what countries are involved, or show videos of former clinical trial participants describing their experience. Incorporate social incentives,{7} such as involving the caregiver in discussions to help provide additional encouragement and support throughout the study.
Attitudes	How do I feel about all that is required of me in the study?	Attitude may extend beyond self-efficacy, as it is shaped by beliefs and factors associated with a behavior.{5} For example, if a patient dislikes sitting in traffic and lives far away, he or she may be less likely to attend required visits. It can be valuable to measure attitudes at the beginning, and throughout the study,{5} via questionnaire, to determine how attitudes are shifting around the required study activities and to determine if intervention may be required. Explaining why patients need to do certain things can reduce perception of just being told what to do, decreasing resistance to change.{6}
Motivation	Do I have desire? Is it compelling, helpful, or interesting?	Get to know your patient to understand what drives them—intrinsically and extrinsically. {1,3} For those who are extrinsically motivated, praise them for completing compliance-based activities. You can also leverage reminder services to pre-program “why” and “affirmation” messages, so patients receive this reinforcement throughout the study. Help them feel a sense of accomplishment by showing progress in the study, or by conveying how their participation is helping research and may potentially help others like them, to motivate those who may be driven by altruism. This may be especially valuable when asking patients to participate in “extra” activities, like consenting to additional biopsies that aren’t required for trial participation.
Intentions	Am I committed to following through?	Intentions are largely driven by a combination of motivation, attitude, and subjective norms.{1} Work closely with patients to understand their intentions to follow through with study requirements. Use specific examples and outline a concrete plan of how they will achieve them.
Triggers/ Cues	How will I remember to do these things, or when I need to do them?	Incorporate nudges that are anchored to a desired action,{3} such as sending reminders near or at the time patients need to complete a diary entry. As habits begin to form, patients may need fewer prompts, as constant reminders may create message fatigue. Shifting to targeted behavior-based notifications can help reduce this. For example, if patients are required to wear an activity monitor, only send targeted messages to those who are at risk of dropping below the required wear threshold.

 

While addressing these elements independently has value, it’s sometimes helpful to incorporate them into an overarching strategy using the model to drive a more holistic strategy around engaging your patients. The model(s) you reference may vary based on needs{9}:

• Stage of the study (recruitment vs. in-trial)
• Protocol requirements (complexity and number of assessments)
• Study aims (prevention vs. treatment trials)
• Patient population (age, gender, indication)
• Region and associated culture (social constructs, motivations)
• Successful previous applications of the model in similar studies/populations

While there is an extensive list of behavioral models that could be leveraged, Table 2 presents a few worth considering. The sections following this table provide descriptions of the four models.

 

Table 2: Constructs Found in Behavioral Models to Consider for Patient Engagement

		Extended Parallel Process Model{4}	Theory of Planned Behavior{5}	Fogg Behavior Model{3}	The Information–Motivation–Behavioral Skills Model{1}
Constructs	Perceived Risk/Threat	✓
	Knowledge				✓
	Skills/Ability			✓	✓
	Perceived Self-Efficacy/Behavioral Control	✓	✓
	Response-Efficacy	✓
	Subjective Norms		✓
	Attitude		✓
	Motivation			✓	✓
	Intention		✓
	Triggers/Cues			✓

 

Extended Parallel Process Model

The Extended Parallel Process model is intended to predict how people will respond to fear of a risk, given their level of perceived efficacy. Risk is inclusive of perceived susceptibility and severity, while efficacy includes self and outcome efficacy.

The model predicts that if people have high perceived risk and low efficacy, they will begin fear control behaviors by avoiding or denying the issue. Those with high perceived risk, but even higher perceived efficacy, are motivated to begin danger control activities by taking action to reduce the risk.{4} This model is more often used to drive fear appeals by creating just enough fear to motivate someone to act.{10}

This may be applied to communication strategies for those with rare diseases or patients on their last line of treatment, who may tend to have lower risk perceptions to trial participation, given the alternative.{10} With that said, perceived outcome efficacy may be low in these populations, but potentially still higher than the perceived risk of the study, thus motivating them to participate.

Furthermore, within the trial, it can be valuable to heavily reward these patients for completing self-efficacy building activities, such as basic compliance requirements. In this context, patients in a Duchenne muscular dystrophy trial may experience learned helplessness, or a sense of powerlessness. It may be that physical therapy does not lead to progress for the patient, and no matter how hard the patient works, physical functions continue to decline.

When patients don’t see a positive outcome from all their hard work, they may be more inclined to give up.{11} This is where focusing on compliance, such as rewarding patients by doing something within their control (e.g., wearing their activity sensor), can help to reinforce self-efficacy. It’s also valuable to continue reminding them of how their participation is helping researchers find a treatment for others like them (outcome efficacy). This may motivate patients to stick with it, even if they’re not feeling better.

Theory of Planned Behavior

This theory posits that a person’s attitude, social perceptions, and self-efficacy can drive them to act.{5} A potential patient population to apply this model to would be children and adolescents, who are typically more susceptible to and impacted by social influences than other age groups.

Although children are required to consent to participate in a study, consent is also necessary from a caregiver. This could serve as an opportunity to emphasize to the patient that while he or she needs parental consent, the child also has a say in participation.

Targeting a child’s attitude toward the behaviors (visits, medication, tracking) may be more effective than just instructing them on what to do.{6} With that said, education and assigning achievable tasks, with support from caregivers, can help build confidence early on in the trial.

As school-age children grow, they are increasingly more influenced by their peers and social environment.{2} Leveraging social influences,{7} such as describing basic characteristics about others in the trial (e.g., age and country), highlighting celebrities with the same or similar condition, or showcasing commercials or PSAs, helps reduce the stigma. This can help make trial participation feel more mainstream and “socially acceptable.” Caregivers can also serve as a source of social reward when young children feel like they are pleasing their parents by doing what is asked of them.

However, it’s important to use social strategies carefully. While positive reinforcement from a social network for high compliance can drive positive behaviors, it might prompt certain patients to feign compliance activities to please those in their social network. Therefore, one must find the right balance of rewarding compliance, while also rewarding honesty.

Fogg Behavior Model

The Fogg Behavior Model states that for people to be successful in performing a behavior, they need to be motivated, have the skills and ability to perform the behavior, and be prompted by a trigger. Strong presence of all three constructs equates to a higher likelihood of success. While a trigger must always be involved, motivation and skills do not both necessarily need to be high, so long as one of these constructs is compelling enough.{3}

If someone is highly motivated, but has minimum skills to perform a task, motivation in itself might drive the person to acquire the necessary abilities. On the flip side, if something is easy, people might just do it.{3} Take the case of store clerks who ask if you would like to donate your change or a certain small dollar amount to charity upon checkout. On your own, you may never have been motivated to donate, but in this instance, it is so easy that you agree when prompted.

Motivation can be quite elusive, given it is so unique to the individual. According to the Center for Information and Study of Clinical Research Participation, motivators for participating in trials are to help advance science and treatment of disease/condition, help others (altruism, or family), receive compensation, and to obtain better treatment. The top burden impacting ability is traveling to the study clinic. In fact, the top voluntary reason for leaving a study is the location of the study center.{12}

In clinical research, sponsors and vendors are working on various ways to remove friction and address the “time,” “money,” and “effort” abilities by presenting opportunities for patients to participate in studies more easily, whether by introducing remote visits, offering smartphone apps, or even providing childcare.

The Fogg Behavior Model may be applied to patients on two sides of the spectrum. For example, patients who are seeking last-line oncology treatment may be more motivated to participate, comply, and stick with a trial. Healthy participants, like in the case of certain vaccine trials, may not be highly motivated, but if it’s easy for them to do, they may be more likely to participate. This is especially the case if some extrinsic motivation can be addressed through compensation, including compensation associated with compliance activities.

In these scenarios, both groups need a prompt to participate, whether it be from a doctor for the highly motivated, or by passively being exposed to an online ad that makes it easy to sign up. Both groups would benefit from in-trial patient engagement strategies to trigger an action, like reminders anchored to the timing of an expected behavior (e.g., filling out a diary) and apps that prompt a call-to-action on a patient’s phone, which 90% say they “frequently” carry with them.{13}

The Information–Motivation–Behavioral Skills Model

This model suggests that factors that influence behavior include knowledge about the behavior, motivation to take action, and behavioral skills necessary to complete it. Not only must information and motivation be tied to skills, they must also link to the behavior change outcome.{1}

Information can motivate.{2} An obesity study protocol may prescribe that patients complete specific health-behavior regimens, such as exercising and eating right. Patients need to be informed on what it means to eat right, and what type and level of exercise is most appropriate. They must also understand why and how this helps in the study. These guidelines, regardless of what arm of the study the patient is in, can reduce the patient’s weight, therefore motivating them to comply.

Behavioral skills can be reinforced by providing tools such as pedometers or food scales,{1} and “tiny habit” exercise reminders can be added into a patient’s daily routine (taking the stairs instead of elevator, walking to a colleague’s desk rather than messaging them, etc.).

Research has shown that patients immediately forget 40% to 80% of the medical information they receive, and about half of what is retained is incorrect.{14} Therefore, these details must be made available to patients after their visits. Critical study information, presented in various formats, such as combinations of text, images, and interactive modules, can also be conveyed through the use of technology to target the diverse learning styles of patients.{14} Questionnaires can evaluate comprehension, and the results can be then used for targeted training to ensure knowledge is maintained.

Key Considerations

The common thread in these models is that many of the constructs can be leveraged to influence patients’ way of thinking to promote specific behaviors. However, in some studies in which the endpoint is to measure changes in thinking, such as psychological symptoms of depression, initiating tactics to improve self-efficacy may be considered as interventional, and therefore needs to be navigated carefully.

Patients who are motivated by knowing they are helping to potentially take a drug to market may be inclined to report more positive results about how they are feeling, in hopes that it can help facilitate that process. While placebos can help offset this, it’s important to reinforce to patients the value of answering honestly and to describe the benefits of being truthful when there are issues.

We also know that participation may be driven by motivation to potentially treat the disease. Again, it’s critical to remind patients the intention of clinical research is to answer questions, and while the goal is to identify an intervention that is safe and effective, that is not the sole intention of the study.

Additionally, one must be careful in how fear appeals are leveraged to initiate behaviors. While it is important to emphasize to patients the importance and criticality of preventing and reporting adverse events, conveying such information must be objective and balanced relative to the risk. The key is to not overly and unduly communicate risk without emphasizing mitigation and escalation strategies.

To summarize, constructs and behavioral models are not prescriptive, but are intended to be referenced alone or in combination to help guide patients throughout the trial, and to help predict and reinforce or mitigate certain behaviors. So, whatever you opt for, approaching your patient engagement and retention strategies with a plan in mind can help improve the trial effectiveness and patient satisfaction. Why not give it a try?

References

1. Amico KR, Mugavero M, Krousel-Wood MA, Bosworth HB, Merlin JS. 2017. Advantages to using social-behavioral models of medication adherence in research and practice. J Gen Int Med 33:207–15. https://www.ncbi.nlm.nih.gov/pubmed/29204968
2. World Health Organization. Adherence to long-term therapies—evidence for action. https://apps.who.int/medicinedocs/pdf/s4883e/s4883e.pdf
3. Fogg BJ. 2009. A behavior model for persuasive design. Persuasive ’09. https://dl.acm.org/doi/10.1145/1541948.1541999
4. Glanz K, Rimer BK, Lewis FM. 2002. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco: Jossey-Bass.
5. The World Bank. Communication for Governance and Accountability Program. Theories of Behavior Change. http://documents.worldbank.org/curated/en/456261468164982535/Theories-of-behavior-change
6. Britt E, Hudson SM, Blampied NM. 2004. Motivational interviewing in health settings: a review. Pat Ed Counsel 53:147–55. https://www.ncbi.nlm.nih.gov/pubmed/15140454
7. VanEpps EM, Volpp KG, and Halpern SD. 2016. A nudge toward participation: improving clinical trial enrollment with behavioral economics. Sci Transl Med 8(348):348fs13. https://stm.sciencemag.org/content/8/348/348fs13.full
8. Czaja SJ, Charness N, Fisk AD, Hertzog C, Nair SN, Rogers WA, Sharit J. 2006. Factors predicting the use of technology: findings from the Center for Research and Education on Aging and Technology Enhancement (CREATE). Psychology and Aging 21(2):333–52. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1524856/
9. National Institutes of Health. Office of Behavioral and Social Sciences Research. Social and behavioral theories. https://obssr.od.nih.gov/wp-content/uploads/2016/05/Social-and-Behavioral-Theories.pdf
10. Morrow GR, Hickok JT, Burish TG. 1994. Behavioral aspects of clinical trials: an integrated framework from behavior theory. Cancer 74(9)Suppl:2676–82. https://acsjournals.onlinelibrary.wiley.com/doi/pdf/10.1002/cncr.1994.2820740918
11. Hendriksen J, Hendriksen R, Kuijer J, Vroom E. Duchenne Parent Project. The psychology of Duchenne muscular dystrophy. https://www.parentprojectmd.org/wp-content/uploads/2018/04/Psychology_of_Duchenne.pdf
12. Center for Information and Study on Clinical Research Participation. 2019. New patient engagement insights from the 2019 CISCRP Perceptions & Insights Study. https://www.ciscrp.org/new-patient-engagement-insights-from-the-2019-ciscrp-perceptions-insights-study/
13. Pew Research Center on Internet and Technology. Americans’ Views on Mobile Etiquette (Chapter 1: Always on Connectivity). https://www.pewresearch.org/internet/2015/08/26/chapter-1-always-on-connectivity/
14. Kessels RPC. 2003. Patients’ memory for medical information. J Royal Soc Med 96(5):219–22. doi:10.1258/jrsm.96.5.219. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC539473/

Mindy Gruba, MPH, is a Senior Product Manager with Signant Health.