Abstract
In March 2020, most institutions of higher education (IHEs) closed their campuses and transitioned to online learning in the wake of the coronavirus disease 2019 (COVID-19) pandemic. With the COVID-19 pandemic continuing for the foreseeable future, IHEs had the difficult task of determining their plans for the fall semester and the remainder of the 2020-2021 academic year. As of January 4, 2021, more than 20.5 million COVID-19 cases had been confirmed and more than 350 000 deaths were directly attributed to COVID-19 in the United States, 1 and universities were considered one of the highest-risk settings for transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes COVID-19. 2 As of December 11, 2020, more than 397 000 cases of COVID-19 had been confirmed on college campuses nationally, 3 and COVID-19 cases on college campuses have been associated with community spread. 4 Young adults who do not have comorbidities have been found to have prolonged illness after COVID-19 diagnosis. 5 With 20 million enrolled students and an additional 4 million faculty and staff members in more than 6000 postsecondary institutions across the United States, 6,7 IHEs have made and continue to make decisions about COVID-19 prevention that will have widespread implications for both people affiliated with IHEs and communities at large.
About half of IHEs reopened fully in person, primarily in person, or via a hybrid in-person/virtual model in fall 2020 to give students a campus-based college experience 8 but with numerous stipulations in place to protect the campus community from SARS-CoV-2 transmission. Provisions have included de-densifying residence halls, remodeling classrooms, requiring daily health assessments, providing routine testing, and having students sign social contracts (also called pledges, agreements, and compacts) agreeing to abide by behavioral measures to reduce the risk of acquiring and transmitting SARS-CoV-2. 9 -13 For the latter, the publicly available social contracts commonly include actions such as self-reporting COVID-19–related symptoms, staying home if ill, washing hands or using hand sanitizer regularly, wearing face coverings, engaging in social distancing, and avoiding large social gatherings. 14 -17 Students are usually required to sign the contract before being allowed back on campus. Although some contracts state that noncompliance may lead to disciplinary action, most contracts lack explicitly stated enforcement measures.
Although the safety of reopening IHEs has been divisive among people associated with IHEs and the community at large, the premise that students will be motivated to modify their COVID-19 risk behavior based on a social contract has been particularly controversial. 18,19 Some people claim that contracts will remind students of campus standards and give them the opportunity to do the right thing for themselves and their community to safely continue their campus-based experience. 9,10 Other people cite increasingly common reports of college students engaging in activities that facilitate SARS-CoV-2 transmission, particularly congregating for social events, that have led to COVID-19 outbreaks on campuses. 20 -23 Research on adolescent development supports these concerns. The traditional college age range (ie, 18-24) is a time of heightened risk-taking behavior, particularly because young people are drawn to short-term rewards. 24 In fact, the number of COVID-19 cases increased 55% among young adults aged 18-22 nationally from August 2 through September 5, 2020, during which time many IHEs reopened in person. 25 In a summer 2020 survey of more than 7000 faculty members, staff members, and graduate students at a university planning to institute a social contract, 67% of respondents expressed doubts that it would be effective. 19
Many IHEs that have reopened in person are heeding national COVID-19 prevention guidelines that primarily focus on the campus environment. 26 -28 Although the environmental changes suggested in these guidelines are grounded in scientific evidence, questions remain about how to most effectively influence students’ COVID-19 risk behavior 18,29 and whether social contracts are the most advantageous way to do so. Public health interventions should be evidence based, 30 and given the rising number of COVID-19 cases on college campuses, an “urgent need to implement effective mitigation strategies” 31 is needed. Fortunately, considerable evidence may inform the use of social contracts for college students in the era of COVID-19. In this commentary, we synthesize the literature on social contracts and provide evidence-based recommendations for IHEs about their use during the 2020-2021 academic year and beyond. We contend that relying on social contracts, particularly in their current form, will not safeguard campuses from COVID-19.
Behavioral Contracts: A Literature Synthesis
Having people sign contracts as a tool to enable behavioral change is a longstanding practice. Usually referred to as “behavioral contracts” in the literature, having people sign contracts originated as a therapeutic technique in the 1960s and soon became an established and recommended strategy in therapy settings. 32,33 According to Epstein and Wing, 32 “a behavioral contract involves a written agreement specifying operationally defined behaviors and contingencies designed to promote behavior change.” Other scholars offer similar definitions and consensus that behavioral contracts are an accepted and flexible clinical tool. 33 -35 Behavioral contracts are intended to improve people’s adherence to behavior-change interventions and to bridge the gap in time between a person’s initiation of a new health behavior and the behavior becoming a health habit. 32,36
Behavioral contracts should contain several key components. They should focus on specific and detailed behaviors that have been determined in partnership with contractees and are within the contractees’ ability to perform. They should also include positive reinforcements (or rewards) that depend on fulfilling the contract, along with negative reinforcements (or consequences) if the contract is not fulfilled. Both kinds of contingencies should have motivational value and be quickly executed. In addition, contracts should have a set time frame and a clear way of measuring and monitoring the contract’s progress. 32 -35
An exhaustive review of the literature from the 1960s and 1970s by Epstein and Wing 32 found that behavioral contracts were primarily used to improve smoking cessation, mitigate drug use and abuse, support weight control, and increase attendance at therapy sessions, and that behavioral contracts generally yielded positive results. A similarly timed literature review of 49 experimental studies by Kirschenbaum and Flanery 33 found that behavioral contracts were used in 3 major areas: family/marriage interactions, academic behaviors, and health behaviors. The authors concluded that behavioral contracts consistently facilitated behavior change compared with minimal or no treatment controls; however, behavior change was usually short-lived. They also determined that having contractees participate in the contract development process was especially important in increasing their commitment to behavior change.
Behavioral contracting has persisted as a behavior-change approach with an expanded number of behavioral outcomes in both therapy and medical settings. 34,35,37,38 Research examining the effects of behavioral contracts has produced favorable findings, although many studies have had small sample sizes and lacked longitudinal follow-up. 34 Using a tailored and participatory process, monitoring progress, and ensuring that contracts are not perceived as too rigid, authoritative, or punishing have consistently been identified as essential factors. 35
Despite their clinical origins, behavioral contracts have become an established public health intervention strategy during the past several decades. Abraham and Michie 39 classified behavioral contracts as 1 of 26 distinct behavior-change techniques, defined as “an observable, replicable, and irreducible component of an intervention designed to alter or redirect causal processes that regulate behavior.” 40 Health promotion program planning textbooks now list contracts as a recognized behavior-change method. 30,41 Interventions such as behavioral contracts that are driven by theory are more likely to be effective than intervention strategies that are not theory based. 42 Behavioral contracts are rooted in principles of operant conditioning, which employ rewards and punishments for a behavior to create associations between the behavior and its related positive or negative consequences. 35,39 Behavioral contracting also incorporates several social cognitive theory constructs, including self-control, reinforcement, and self-efficacy. 42
To date, behavioral contracts have been used across diverse public health issues and populations. They have served as stand-alone interventions but are typically one element of multicomponent programs. Behavioral contracts have been used to address physical activity and healthy eating, 38,43 -45 asthma management, 46 modeling healthy behavior, 47 hospital-acquired infections, 48 smoking, 49 self-management of chronic diseases, 50,51 and abstaining from sex until marriage, also referred to as “virginity pledges.” 52 Although the core components of behavioral contracts have been used inconsistently across studies, behavioral contracts have largely been considered beneficial. The one exception is a virginity pledge, which involves a one-time signature with no other provisions, fully incompatible with behavioral contract tenets.
A Cochrane review of 30 randomized controlled trials examined behavioral contracts used in patient/health care provider and health-promotion initiatives in high-income countries. The review found “limited evidence that contracts can potentially contribute to improving adherence, but . . . insufficient evidence from large, good quality studies to routinely recommend contracts for improving adherence to treatment or preventive health regimens.” 36 This conclusion was largely based on individual studies that yielded positive findings but had study design limitations, including a lack of longitudinal follow-up. Behavior change was typically not sustained when it was assessed.
In sum, behavioral contracts have a longstanding history and have been used in therapy, medicine, and public health settings across a wide range of health issues, behaviors, and priority populations. Behavioral contracts are often used to evoke behavior change at the individual level (eg, healthy eating) and to protect the health of people’s interpersonal and community networks (eg, hospital infection spread). They have a defined structure, are theoretically based, and have been deemed a formal behavior-change technique. Individual studies tend to identify behavioral contracts as effective tools for behavior modification; however, reviews have found that in practice, behavioral contracts are implemented variably and behavior change is typically not prolonged. This literature synthesis provides important insights into the use of behavioral contracts for COVID-19 prevention among students at IHEs.
Student Contracts to Prevent SARS-CoV-2 Transmission: Evidence-Based Recommendations
Based on scientific evidence, behavioral contracts may have utility in public health interventions to safeguard the health of individuals and their communities. They will likely be most effective if people are motivated to modify their behavior; specific behavioral goals, rewards, and consequences are tailored to the individual; the contractees’ progress is monitored and they are held accountable for the agreed-upon content; the contract is time limited; and the contract is just one facet of a multicomponent intervention. Yet, even under these ideal circumstances, behavior change driven by behavioral contracts may be short term at best. COVID-19 student contracts, in juxtaposition, are required; contain numerous broad and standardized behaviors decided upon by the institution; and lack contingencies, monitoring, and a clear-cut time frame.
Although some contracts mention enforcement, no explicit information is usually provided about what enforcement would entail, and enforcement is solely punitive (eg, student suspensions at several IHEs in fall 2020 that garnered considerable news coverage). 53 Whereas behavioral contracts are intended to be empowering, 37 COVID-19 student contracts are only restrictive. Enforcement may be particularly important in reducing health-threatening behavior among college students, because enforcement has been found to be a critical component in preventing other problematic health issues, such as tobacco use and high-risk alcohol consumption, at IHEs. 54 Furthermore, although students would ideally be intrinsically motivated to engage in COVID-19 prevention behaviors, behavioral contracts are able to generate external motivation (ie, motivation because of external factors), which can provoke behavior change. 41 However, external motivation cannot occur without contingencies or enforcement, neither of which are included in COVID-19 student contracts. Ultimately, COVID-19 student contracts in their current form do not incorporate any known best practices from the literature or behavior-change theory.
Students’ behavior must also be contextualized within the broader environment in which they reside. 55 Much of students’ college experience takes place outside the classroom. 56 College is a period of psychosocial development, during which students cultivate a sense of belonging and seek social connectedness. 57,58 COVID-19 contract restrictions, such as social distancing, quarantining, and refraining from group gatherings, are antithetical to these core elements of collegiate life. Combined with the COVID fatigue (ie, emotional exhaustion due to COVID-19) that students are likely already experiencing, 59 it would be intelligible that students’ risk/reward ratio may become skewed, especially if the consequences are invisible (ie, students not discerning their role in community spread on- and off-campus). As adolescent researcher Laurence Steinberg 60 stated about the reopening of campuses, “It’s one of those perfect storms—people who are inclined to take risks in a setting that provides ample temptation to do so.”
Furthermore, students put their trust in their campus administrators. If students are told that it is safe to return to campus, it would be unsurprising that students would consequently underestimate the severity of COVID-19 on campus and within the wider community. Students will likely be collectively lambasted if they falter in their COVID-19 preventive behaviors, yet asking students to be on campus while prohibiting their campus experience is a counterintuitive and precarious position in which to put students. Given the many problems with COVID-19 behavioral contracts in their current form, including their composition and inattentiveness to individual and environmental constraints, they will likely be ineffective at producing widespread behavior change on college campuses.
If IHEs still choose to use COVID-19 behavioral contracts, we recommend they be implemented SMARTly, meaning they have:
Student contracts are the only behavioral intervention that most IHEs have publicly communicated in their campus COVID-19 prevention plans. Different or supplemental behavior-change techniques may prove more effective than contracts alone, including bolstering campus norms that promote a culture of care and having popular opinion leaders endorse and model behavior. 62 Commonly used social media platforms can serve as major channels for social norms campaigns and information distribution.
COVID-19 interventions should be behavior-specific and informed by high-quality evaluations. However, students are being asked to perform multiple protective behaviors with minimal guidance from the literature. 63 In the absence of evidence, West and colleagues 63 advise drawing upon behavior-change principles to help identify opportune COVID-19 prevention strategies. Presumably of particular relevance in the collegiate population are tactics that underscore how the benefits of behavior change outweigh the costs, that heighten dissonance, and that offer social support. Techniques should be selected after conducting a methodical assessment of behavioral determinants. 59 It will be especially important to solicit feedback from students representing diverse sectors of the university, because they can offer meaningful insights and applicable solutions for COVID-19 campus policies. 64 Engaging students in planning processes should lead them to become more invested in what materializes, thereby resulting in more optimal outcomes than if students were not engaged in these processes. 30
Yet even under the best of circumstances, although well-executed behavioral contracts and other behavior-change strategies may improve the likelihood of behavior change, they are not a guarantee. Reliance on an increased probability of behavior change for such a serious infectious disease is a perilous proposition. Purdue University President Mitch Daniels, 9 who has spoken favorably about COVID-19 student contracts, wrote, “Nothing makes a more positive difference than personal behavior and responsibility.” Other college administrators and health experts have similarly discussed behavior as the pivotal force in alleviating the COVID-19 crisis. 18,29,65 We concur in principle, but as behavioral scientists, we also recognize how difficult sustained behavioral change can be, particularly in an environment that is not conducive to it. Institutional administrators have exceedingly difficult choices to make as they determine how to best meet the academic mission of their institutions, maintain financial solvency, and protect the campus and surrounding community’s health and well-being during the COVID-19 pandemic. There are no easy answers. But given the possible dire consequences, expecting students to modify their behavior because of an imposed social contract is an insufficient solution.
Footnotes
Acknowledgments
The authors thank J. Ryan Kennedy for his assistance with this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.