Eliminating nonmedical exemptions: a radical shift in how childhood vaccine mandates govern

ABSTRACT

Introduction

Every state in the US has had school vaccine mandates for decades, and all except West Virginia and Mississippi offered nonmedical exemptions (NMEs) in addition to medical exemptions. Several states recently eliminated NMEs, and others have attempted to do so. These efforts are transforming America’s immunization governance.

Areas covered

What we call the ‘mandates & exemptions’ regime of vaccination policy from the 1960s and 1970s functioned to orient parents toward vaccination, but did not coerce or punish them for not vaccinating. The article identifies how policy tweaks in the 2000s – including education requirements and other bureaucratic burdens – delivered enhancements to the ‘mandates & exemptions’ regime. Finally, the paper illustrates how the recent elimination of NMEs, first in California and then in other states, represents a radical transformation of America’s vaccine mandates.

Expert opinion

Today’s ‘unencumbered vaccine mandates’ (mandates without exemptions) directly govern and punish non-vaccination, unlike the ‘mandates & exemption’ regime that aimed to make it harder for parents to avoid vaccination. This kind of policy change introduces new problems for implementation and enforcement, especially within America’s underfunded public health system, and in the context of post-COVID public health political conflicts.

KEYWORDS:

• California
• nonmedical exemptions
• philosophical exemptions
• religious exemptions
• schools
• vaccination
• vaccine hesitancy
• vaccine mandates

1. Introduction

In America, polarized political conflict about COVID-19 vaccines and mandates has dominated recent public and academic discussion about vaccination policy. But as the pandemic wanes, attention is returning to routine childhood vaccinations. Most Americans (90%) still think that the benefits of the measles, mumps and rubella (MMR) vaccine outweigh the risks [1]. However, they are increasingly divided about state-level school enrollment vaccine mandates: Only 57% of Republicans now support school vaccine mandates (79% did in 2016), while Democratic support for mandates is now at 85% (83% in 2016) [1]. Most of this political divide emerged during the COVID era, but political conflict about vaccination policy was already heating up before COVID struck. During the 2010s, states across the US attempted radical changes to their vaccination policies for school entry. Conflict over these policies fractured existing bipartisan alliances behind childhood immunization programs in ways that planted seeds for polarizing COVID conflicts. Recent state-based conflicts over school vaccine mandates are more significant than they may appear to be. We need to understand what has changed in America’s state legislatures when it comes to these policies, if we want to have a better sense of current political conflict.

School entry vaccine mandates have been the lynchpin of America’s vaccine governance since the second half of the 20^th century. These policies have always offered medical exemptions (MEs) for people who cannot vaccinate for medical reasons. However, today’s vaccine mandates mostly date from a period of consensus starting from the 1960s and extending through the 1970s, during which state policymakers across the U.S.A instituted vaccine mandates, usually accompanied by nonmedical exemptions (NMEs). These policymakers designed their states’ mandatory vaccination policies to orient parents toward vaccination. As former Directors of the National Immunization Program (CDC), Walter Orenstein and Alan Hinman, put the point in 1999, ‘The laws act to make immunization a priority’ [2]. Mandates also had the effect of creating pro-vaccination social norms. However, the policies did not directly govern vaccine acceptance by coercing or punishing refusers, because parents who were committed to refusing vaccines could access nonmedical exemptions to allow their unvaccinated children to attend school. These NMEs permitted parents to opt out for religious reasons, or – in some states – for wider philosophical or personal belief objections. We refer to this form of immunization governance as the ‘mandates & exemptions’ regime.

When California policymakers altered their state’s ‘mandates & exemptions’ regime by removing NMEs in 2015, they began to directly govern parents’ vaccination choices. This was a dramatic change in immunization governance. Scholars have focused on the origins of California’s policy change [3,4], its impact on compliance with vaccination [5–7], and its implementation challenges [8,9]. However, there has been little attention to the significant qualitative change in governance that eliminating NMEs enacts. A growing list of other states have subsequently eliminated their religious or personal belief NMEs (including Maine, New York, Washington, and Connecticut), and this kind of policy change has the support of major US physician organizations [10]. Yet these new ‘unencumbered mandates’ – our name for mandates that are shorn of their prior exemption policies – deviate dramatically from America’s previous ‘mandates & exemptions’ regime, which allowed parents who did not want to vaccinate to access exemptions, enabling them to enroll their unvaccinated children in school.

This article uses legislative and policy histories, as well as interviews with health officials and advocates (approved by The University of Western Australia’s Human Ethics Review Committee RA/4/20/5326), to advance three theses about how the elimination of NMEs by American state governments changes immunization governance:

1. Nonmedical exemptions were central to the governance strategy that state and federal public health policymakers sought when they created what we call the ‘mandates & exemptions’ regime in the second half of the last century.

2. Legislative reforms from the 2000s to early 2010s enhanced this regime by using emergent behavioral governance strategies, e.g. ‘nudging,’ to incline parents toward vaccination.

3. When state legislatures eliminate NMEs (instituting ‘unencumbered mandates’), they radically reshape existing forms of vaccination governance, and place new pressures on implementation and enforcement apparatuses that were created for a qualitatively different purpose.

4. Political resistance is likely to arise against new coercive measures in childhood vaccine policy, considering recent disputes over COVID pandemic control measures.

2. Vaccine mandates address systems failings

America’s political leaders have long relied upon vaccine mandates to encourage the population’s uptake of vaccines. Governing officials in many American cities and states employed mandates for smallpox vaccines from the 19^th century onwards, but by the end of World War II, a short-term shift to vaccine voluntarism was underway [11–13]. It was in this context of voluntary vaccination that polio vaccines were introduced, and in the 1960s, US federal health officials made disease eradication a top goal [14]. State and national actors mobilized funding and developed policies to promote uptake of newly available vaccines, especially for measles [15]. However, efforts to promote voluntary vaccination by providing information for parents did not achieve desired coverage rates [16]. Officials began to reflect that they may have ‘over-estimated’ potential vaccine acceptance, as the chief of epidemiology at the Centers for Disease Control, Alexander Langmuir, put the point in 1960 [14]. When measles eradication campaigns stalled due to wavering federal financial support, governments and advocates turned to schools as sites to govern vaccination. The idea was that parents were ‘basically willing,’ but that they needed to be socialized into giving their children the newly available vaccines, and that mandates would help to boost compliance [13].

Officials’ efforts to promote vaccination during this period paid little attention to the complex reasons why people might not vaccinate. Advocates of vaccine mandates for school enrollment were especially disinterested in those who might deliberately refuse. Instead, the school vaccine mandates of the 1960s and 1970s were motivated by a recognition that the public health system had failed to reach everybody with vaccines, or to sufficiently persuade people to accept them [14–16]. This is a very different perspective from the supporters of NME abolition today, who seek to solve a perceived problem of vaccine refusal by changing the behavior of noncompliant parents. (For example, in 2015, the California chapter of the American Academy of Pediatrics wrote to the Senate’s Health Committee that ‘an increase in the number who opt out from getting their children vaccinated’ can ‘put everyone at risk’ [16]). Without consistent funding to support vaccine advocacy work, voluntary immunization policies would be unlikely to sustain high immunization rates or to ensure sufficient uptake of new vaccines. Governments needed an inexpensive way to promote vaccine uptake in light of institutional and economic limits. School-entry vaccine mandates presented an attractive solution.

Schools are vulnerable to outbreaks, so mandates help to ensure that children are educated and that parents can go to work. Perhaps more importantly, school vaccine mandates can effectively ensure immunization of the broader community, because they govern everyone’s immunization status as they pass through schools. Furthermore, state governments have constitutional authority to govern their communities’ schools, such that it is legally possible to use school mandates for immunization governance. Finally, instead of creating and funding new institutions to deliver or promote vaccines, introducing school vaccine mandates required relatively simple changes to existing school enrollment processes. Reflecting on the logics behind the design of these policies in 1999, former Directors of the National Immunization Program (CDC), Walter Orenstein and Alan Hinman [2], noted that:

[S]chool laws establish a system for immunization, a system that works year in and year out, regardless of political interest, media coverage, changing budget situations, and the absence of vaccine-preventable disease outbreaks to spur interest.

With bipartisan support, state governments began introducing new vaccine mandates in the 1960s and 1970s. The CDC distributed sample legislation, and its Director also contacted each state’s governor individually to press the point [17,18]. In 1968, only half of US states had policies requiring at least one vaccine for school entry, but by 1974, 40 states had vaccine mandates, and by 1981 every state did [19].

School entry vaccine mandates ensured vaccination by making it a condition for school access, rather than by investing in vaccine provision. The federal government later created the Vaccines for Children (VFC, 1993) program to help poorer children get vaccines, and the Affordable Care Act (ACA, 2010) now covers the cost of vaccines for privately insured people. However, these policies came long after the school vaccine mandates of the 1960s and 1970s. Furthermore, VFC and ACA provide indirect support for minimal public health provision, compared to the direct comprehensive public health provided by countries like the United Kingdom [20]. Therefore, America’s vaccine mandates in the 1960s and 1970s did not depend on social trust built through public healthcare services, but rather avoided widespread resistance by allowing opt-outs based on rules.

3. Nonmedical exemptions prevented coercion

Advocates for school vaccine mandates generally did not envision these policies to be coercive, but instead thought that mandates would prompt vaccinations. For example, CDC leader Alan Hinman stated in 1979 that ‘some additional stimulus is often needed to provoke action on the part of a basically interested person who has many other concerns competing for attention’ [21]. Officials frequently disagreed about whether coercive mandates could be ethically justified [22], and the inclusion of nonmedical exemptions ultimately circumvented this disagreement, building consensus around the policy tool. Accordingly, America’s modern vaccine mandates either came preloaded with nonmedical exemptions, as was the case in California [23], or these exemptions were added once governments started enforcing mandates and faced community resistance as children were excluded from school [11,19,22]. Some legislators incorporated exemptions from preexisting laws, including those that sanctioned religious objections, but in other states parents or political actors extracted them through political conflict. However, it is important to consider this conflict in light of the overarching purpose of vaccine mandates at the time (activation devices, not punishments for refusers), and hence policymakers’ small appetite to render them controversial. (School-based political conflicts, such as those related to desegregation, were common in the 1960s and 1970s. Policymakers at that time surely aimed to avoid similar conflicts about new vaccine mandates.) There is no available comprehensive history of how each state’s NMEs exemptions were instituted (the field needs more policy histories like that of Kuo and Conis on California [23]), but it is likely that accommodations for religious beliefs in particular reflected policymakers’ aversion to conflict on religious liberty grounds. The fact that few, small religious sects taught against vaccination may have been further reason to embrace religious exemptions in particular. While each state had their own local political, legal, and historical reasons for designing and implementing the ‘mandate & exemption’ policy [11], policymakers across the country likely anticipated that very few people would request nonmedical exemptions.

One way to understand vaccine mandates is through the concept of ‘selectivity’, which we have previously defined as the degree to which a mandate allows for exceptions or alternatives. Selectivity affects how strongly a mandate influences parents’ decisions to vaccinate their children (or not). One form of selectivity is by rule, which means that there are official criteria and procedures for obtaining exemptions from vaccination, either for medical or nonmedical reasons. These exemptions are publicly documented and transparent [24]. By establishing a ‘mandate & exemption’ policy in almost every US state, policymakers were laying out a clear pathway for selectivity by rule that made space for nonmedical refusal. This kind of policy has been in place for a long time, even though it has changed over time (for example, as courts or officials expanded the scope of religious exemptions) and even though some parents abuse the system (for example, by claiming religious objections that they do not really have). Selectivity by rule makes mandates less salient (less likely to produce the outcome of vaccination) compared to having no exemptions available, but it does so in ways that are easy to monitor and measure. In this paper, we mainly focus on personal belief and religious exemptions as two types of rule-based selectivity that operate on nonmedical grounds, because both have similar functions in enabling vaccine refusal. However, we also discuss their differences in our Expert Review section, with special focus on political and legal issues surrounding religious exemptions.

Only two US states, West Virginia and Mississippi, had no NMEs in recent decades. West Virginia never offered NMEs; Mississippi’s 1960 religious exemption was struck down by the state’s Supreme Court in 1979 [19], although a recent judgment has seemingly reintroduced it, as we discuss later [25]. These outliers notwithstanding, the ubiquity of NMEs across American states constituted what we call ‘the great compromise’ of US immunization policy: the state requires vaccination, most people comply, a few request the available nonmedical exemptions, and the exempted people stay quiet and do not encourage the rest of society to refuse vaccinations. All parties must play their roles for this policy to work, and they did from the introduction of the ‘mandates & exemptions’ regime into the new millennium. But vaccine hesitancy and refusal in the 2000s prompted state governments to revise their exemption policies.

4. Vaccine refusal prompts clever hacks for failing mandates

The early years of the 21^st century witnessed increased skepticism about vaccine safety and efficacy [26,27]. Nonmedical exemption rates began rising in many US states [28,29]. Some refusers were becoming increasingly vocal, and some Republican state legislators appealed to this constituency by attempting to introduce wider exemption categories or make exemptions easier to access. These efforts were usually ineffective, but they succeeded in Arizona, Arkansas, and Texas [30–33]. Some states’ religious exemptions had also been expanded or made easier to access given that the 1^st Amendment prohibits government from privileging one form of religious belief over another, or becoming the arbiter of what a church teaches. Churches sometimes provide only general moral guidance, and the principles of moral theology can appear to pull in different directions. The possibility of intradenominational ambiguity and disagreement led the U.S. Supreme Court – in Thomas v. Review Board 405 U.S. 707 (1981) – to focus on private religious convictions, rather than denominational membership, when determining whether someone’s objections qualified as religious. This expansive conception of a ‘religious’ basis for exemption contributed to the ease with which people across the United States could access religious exemptions.

While vaccine-refusing parents were advocating for the state to make it easier to send unvaccinated children to school, public health leaders and their political allies responded to disease outbreaks by pushing for policy changes in the opposite direction. Outbreaks of pertussis, mumps, and measles illustrated that some communities had insufficient vaccine coverage. For example, Lillvis and colleagues record that early 21^st-century outbreaks of pertussis in Washington and California were concerning for policymakers, and that immunization supporters in those states were unhappy that their comparatively high exemption rates were attracting attention. These actors began pushing to make their states’ vaccine mandates harder to escape, and some turned to the behavioral sciences for new policy tools [34]. The introduction of the resultant policies initiated a new governance phase for vaccine mandates, which Lillvis and colleagues call the ‘West Coast Period of Exemption Contraction’, though it was not limited to states on the West Coast [34].

Researchers’ comparative analysis of the effects of different state-level vaccination policies contributed to policymakers’ appetite for tweaking exemption policies. The academic literature reported higher rates of NMEs in states where NMEs were easier to receive [35–37]. Researchers also demonstrated that these states were more likely to have disease outbreaks [36]. As it became evident that access to exemptions influenced parents’ behavior, legislators incorporated an ascendant public policy instrument into immunization governance: ‘nudging.’ Someone who (re)structures our ‘choice architecture’ can nudge us toward a particular choice without depriving us of our liberty to choose otherwise [38]. Nudges can be used in settings where the population might resist a more coercive approach, and they can cause substantial behavior change without large investments of state resources [39]. We do not suggest that all policymakers who sought to limit vaccine mandate exemption policies during this period were explicitly embracing behavioral governance and nudges. Rather, in their pursuit of ‘what works’, they were applying lessons about how to best redesign exemption policies to promote vaccine uptake.

Indeed, the American public officials who designed the ‘mandates & exemptions’ regime had already been utilizing a ‘nudge’, even though they would not have recognized this terminology. The mandates that they designed inclined parents toward vaccination, but NMEs preserved those parents’ liberty to refuse vaccination. Now, with vaccine refusal on the rise, policymakers and researchers were recognizing that the ‘mandates & exemptions’ regime often did not nudge enough. Some parents were obtaining NMEs simply because that was easier than getting their children vaccinated. Policymakers in several states attempted to correct this problem by making it more difficult for parents to access NMEs, often by requiring parents to first receive counseling or education about vaccines and vaccine-preventable diseases. Legislation to require this kind of expert intervention was passed in Washington (2009) and California (2012), while Michigan’s Department of Health and Human Services brought in a version of this policy without legislative amendment [40].

Education requirements proved to be a useful barrier to parents obtaining NMEs, even if they did not change many parents’ minds. For example, beginning in 2015, Michigan’s state government required exemption-seekers to attend education sessions at health department offices [41]. Researchers in Oakland County tracked the vaccination status of children who received nonmedical exemptions during the 2017–18 school year (n = 4,098). Other than parents who were following ‘alternative schedules’, very few parents (7.7%) subsequently agreed to have their children receive vaccines they refused at the session [42]. Michigan’s statewide NME rate nevertheless declined by 35% in its first year, from 5.2% to 3.4% [43]. The burden of attending an education session seems to have nudged many parents to vaccinate, even if the education usually did not change anyone’s mind.

Washington State implemented a similar strategy from 2011, but parents could be counseled by any vaccine provider, who would then sign the exemption form [40]. Legislators in other states pushed for similar policies [34]. In 2012, California Assemblyman Dr Richard Pan led the passage of a copy-cat bill.

5. The failure narrative and the case for ‘Unencumbered Mandates’

Dr Pan and his supporters thought they were importing a successful policy from Washington State, and that they would also be able to optimize the way that mandates & exemptions nudged parents to vaccinate in California. But by the time California’s version came into effect in 2014, its advocates had already started to sour on it, in light of perceived diminishing returns from Washington State’s policy, which did not continue to deliver further immunization rate increases after its first year [44]. However, Pan and his allies had no plans to make California’s mandates more restrictive (2019 interviews with Catherine Flores Martin, Kris Calvin, and Dr Richard Pan; unreferenced, see ‘Notes’). Rather, Dr Pan was planning a new bill that would require schools to publicize their vaccine coverage data, seeking to increase data transparency so that pro-vaccine parents’ ‘choice and voice’ could drive up vaccine coverage [45], a strategy that legislators in some other states were also pursuing during this period [40].

Yet on 19 February 2015, Dr Pan (now a state senator) and Senator Ben Allen introduced Senate Bill 277 to eliminate NMEs in California. The law went into effect on 1 July 2016.

The reasons for this policy change have been elaborated elsewhere [3,4], and it is beyond the scope of this article to address them, though the Disneyland Measles Outbreak of 2014 played a catalyzing role.

We contend that California’s abolition of NMEs heralded the decline of the short-lived period of optimism about the benefits of behavioral governance strategies (like nudging) for immunization policy. This was particularly the case for elected officials and influential medical advocacy organizations, even as some scholars still support these policies [46]. As we have noted, whether or not state policymakers had explicitly employed the philosophy and terminology of behavioral governance, they had used its methods when they sought to make NMEs harder to access. By contrast, removing NMEs altogether ushered in a new era of what we have called ‘unencumbered mandates’, in which state governments repurposed their vaccine mandates to coerce parents to vaccinate and punish them for noncompliance.

These new mandates are coercive because, for at least some vaccine-refusing parents, the threat of preventing unvaccinated children from attending daycare or school is enough to compel parents to vaccinate. (It is an ‘offer they cannot refuse.’) California’s parent activists who worked to remove NMEs believed that introducing this kind of coercion to their state’s mandates would restore a ‘pure’ form of mandate policy (2019 interview with Leah Russin; unreferenced, see ‘Notes’). But since the original function of these mandates was – in today’s behavioral governance language – to activate choice and to socially norm vaccination, the elimination of NMEs did not revert mandates to a prior ‘pure’ form. The removal of NMEs changed the nature of vaccine mandates entirely.

Coercive ‘unencumbered mandates’ may appear to policymakers and to the pro-vaccine public as the only tool left to facilitate vaccine uptake, especially when appeals to community solidarity seem to have little force [47]. The fact that America invests so little in public health and vaccine promotion is a structural problem with the country’s vaccine governance. America depends on mandates and leaves the costs and promotion privatized to insurance companies and individual clinicians. This weakens the capacity, the understanding, the motivation, and the tools policymakers have to promote voluntary vaccination. Indeed, we contend that this lack of investment was a driving factor behind the original ‘nudge’ of the ‘mandates & exemptions’ regime: in 1999 Orenstein and Hinman sagely noted that ‘mandatory immunization may not be needed or appropriate’ in societies ‘with healthcare systems that cover the entire population and stress [disease] prevention’ [2].

But, paradoxically, we believe that the institutional shortcomings of America’s public health system will also undercut attempts to enforce ‘unencumbered’ vaccine mandates. As noted above, in implementing and maintaining the ‘mandates & exemption’ model of vaccine governance, American governments have relied largely on what we now label nudges to govern vaccine uptake. They have not facilitated people’s long-term interactions with well-run public health systems – interactions which can cultivate public trust and support for vaccination [48]. Countries with high trust in civil servants have the highest levels of vaccine confidence [49], and providing successful and functional public services, like vaccination programs, is a prime way to build that trust. For example, one reason why the United Kingdom has been able to have a successful voluntary vaccination system is because of its robust public service provision and outreach efforts [20,48,49]. Dorothy Porter contrasts this kind of universal and unconditional public provision of health goods with the US model of ‘conditional citizenship,’ where healthcare is conditional on social obedience (such as regular employment), a format that policymakers have also applied to children’s health by making school attendance conditional on either vaccination or formal exemption [50]. If America’s governments had governed vaccination through unconditional service provision and outreach, coercion might not even be necessary, although it is possible that investment in such efforts might still have failed to attain sufficient vaccination coverage. After all, many political communities across the globe have recently made their childhood vaccination systems more coercive [4,51]. But having invested in that relationship between people and government, America would also be in a better place to make coercion successful [52].

Employed on their own, then, ‘unencumbered mandates’ are blunt instruments for governing vaccine acceptance, and may produce unintended and undesirable consequences [53,54]. They make some parents vaccinate, but they inspire others to pursue workarounds, or to accept penalties and keep their children unvaccinated. This is evident in California’s implementation problems [8,9]. For example, parents gamed the medical exemptions system so extensively that legislators had to crack down on clinician fraud [55].

Scholars have argued that NMEs function as a metaphorical valve to release the pressure that committed vaccine refusers will otherwise generate within a mandatory vaccination system [56]. Without NMEs, this pressure moves elsewhere. The medical exemption fraud in California demonstrated how committed refusers will continue to use whatever rule-based selectivity is available to them, but may break those rules – in that case, conspiring with health professionals who offer support for economic or ideological reasons. When Vermont altered its NMEs in 2016 to eliminate personal belief exemptions and retain only religious exemptions, researchers observed a replacement effect through a 2.2% increase in the latter [57] that likely also arose from considerable parent dishonesty. Likewise, when Washington State removed its PBE for the MMR vaccine only in 2019, researchers subsequently observed a 367% increase in religious exemptions despite the state’s low levels of religiosity [58].

The removal of NMEs altogether pushes the pressure of resistance elsewhere in the system, forcing the other kind of selectivity we previously identified, which is not rules-based but ad hoc and personal [24]. Enforcement agents can choose not to enforce vaccine mandates upon resistant parents, either because they possess this discretionary capacity at a local level, or because they work within a broader system that provides disincentives to enforce. Both can be true within the school system, where excluding unvaccinated children places a heavy burden on bureaucrats (in health and education sectors), which can limit effective implementation [9]. Even in Michigan – where the mandatory education requirement for accessing NMEs is designed only to orient parents away from exemptions, rather than to deny access to them – some school staff resist promoting immunization in their interactions with families [59]. American and Australian educators have identified a profound mismatch between their commitment to the education of all youth and their role in vaccine mandate enforcement, which includes excluding unvaccinated children from school [24,59–62]. Against this background, ‘unencumbered mandates’ may falter due to enforcement failures, as Delamater and colleagues found when unvaccinated California children continued to be enrolled in schools despite the abolition of NMEs in that state [63]. Importantly, this local ad hoc non-enforcement introduces greater inequality into how mandates govern parents’ behaviors. Private schools, for example, may be less willing to exclude children when their financial functioning is dependent upon high enrollments. This pressure may also apply in some public schools more than others. And because non-enforcement is discretionary rather than rules-based, parents who possess higher levels of economic and social capital may prove more capable of extracting it from beleaguered school officials.

Governments need to invest in additional resources and institutions to make ‘unencumbered mandates’ work. Consider that shortly after California abolished NMEs, Australia, Italy and France all made their vaccine mandates more coercive, either by removing NMEs (Australia) or by expanding the number of required vaccinations and reinvigorating enforcement regimes (Italy and France) [51,64]. The governments of these countries paid for public communications campaigns to augment their new policies; they employed persuasion and informational efforts to win the public’s ‘heart and minds’ over to the benefits of vaccination (and, indeed, to the logic of mandating it) [48]. These governments recognized that coercive public policies should be accompanied by expanded efforts to promote voluntary vaccination, so that governments coerce as few people as possible [52]. Crucially, it is not just the general public that benefits from vaccination information/persuasion campaigns and from longer-term efforts to build public trust in vaccination programs. These efforts can also garner the support and buy-in of enforcement agents in health and education bureaucracies described above. For example, despite some reticence to exclude unvaccinated children from education and care, Australian childcare operators identified with a discourse that was also common in government circles that being pro-vaccine meant supporting vaccine mandates that contributed to herd immunity and protect others [62,65]. If American governments do not invest the resources to fight for hearts and minds when it comes to immunization, they limit their capacity to attain higher levels of compliance of parents and enforcement agents, and their coercive immunization policies may not be effective.

States can adopt more flexible measures in altering their NME policies, but even these are not without risk. Washington State abolished its NME only for the measles vaccine mandate in 2019, changing what Attwell and Navin call the ‘scope’ of a mandate (which vaccines are included or excluded [24]). Washington’s strategy responded to the serious threat posed by measles while attempting to avoid the more radical changes to immunization governance that this paper describes [60,61], and was able to deliver a relative increase in MMR coverage rates of 5.4% even with a marked increase in religious exemptions [58]. However, without persuasion campaigns aimed promoting voluntary uptake of all recommended vaccines, such a piecemeal policy risks signaling to parents that only the measles vaccine is important [24].

6. Conclusion

The march of NME abolition has continued. Within less than a decade, five US states have implemented what we call ‘unencumbered mandates’, and legislators in several additional states have unsuccessfully sought to do so. For example, elected officials in Vermont, Maryland, Rhode Island, and North Carolina all attempted to abolish NMEs in 2015 alone [66].

Eliminating NMEs may have seemed like yet another innovative policy tweak in the spirit of the ‘West Coast Period’ of nudging and policy optimization. After all, it only required the removal of an exemption from an existing general rule. But California’s policy change has set in motion a radical reshaping of immunization governance in that state, and in the other states that have followed California’s lead. Mandates have morphed from activation devices and nudges into coercive instruments, and this may play out in undesirable and unexpected ways.

‘Unencumbered mandates’ can succeed only when most people voluntarily vaccinate and when large majorities endorse the use of coercion for the few who continue to refuse. Other countries, such as Australia, have been able to create these ‘success conditions’ for coercive vaccine mandates by using complementary governance tools [67]. In contemporary America, unencumbered mandates are likely to be less effective, less scalable, and more fragile.

7. Expert opinion

For three years, the COVID-19 pandemic has consumed the attention of policymakers and the public. Even as Americans continue to attend to the importance of COVID-19 disease control measures, they should not lose sight of the dramatic changes that have been happening in their country’s governance of routine childhood vaccines. The elimination of NMEs is a substantial policy change that may produce significant outcomes for public health. But perhaps as important are the impacts that NME elimination efforts may have on America’s ability to maintain stable governance of vaccine uptake, especially with the accelerant of political conflicts over COVID-19 public health orders, interventions, and treatments.

Republican backlash against Democrat efforts to eliminate NMEs includes efforts to water down vaccine mandates; Republican politicians are now playing to a committed constituency, with recent Pew research reporting that now only 57% of Republican voters support vaccine requirements for school – down from 79% in 2019 [1]. If political polarization over vaccine mandates extends to other aspects of immunization policy – and there is little reason to think it will not – then some opponents of vaccine mandates may extend their critiques to government efforts to fund, promote, and provide vaccines for the nation’s children.

Religious exemptions to school entry vaccine mandates will be an important area of future focus. Religious exemptions are not markedly different from personal belief exemptions as governance instruments for rules-based selectivity, as we have argued above. Both types of exemptions are accessible – to a greater or lesser degree – to Americans who wish to refuse vaccines for a range of reasons. However, despite their similarity as governance instruments, religious and personal belief exemptions differ at a socio-political level. When religious exemptions are the only NME available, committed refusers will seek them out, as discussed above, and defend their maintenance. In states that offer only religious exemptions, the rates of these exemptions have been rising, even as overall levels of religiosity have diminished in the populace [35,57]. And since many more states have religious exemptions than personal belief exemptions, substantial contestation will likely center on the former in the future. There is reason to be particularly concerned about how these debates will play out, given the importance of religious identities to claims of rights and freedoms within American political history and the present culture wars. These contests may bring actors into vaccine policy debates who would otherwise have little interest in participating, but who will further cleave apart the former bipartisan consensus on the merits of vaccination.

Legislators in Kansas are presently seeking to make that state’s religious exemption (the only NME available) accessible via declaration. This would replace the more onerous current requirement to demonstrate that one’s religion formally opposes vaccination. This intervention builds directly upon a successful insertion of a COVID-19 vaccine religious exemption by declaration in 2021 in Kansas – a measure which also criminalized employers challenging exemptions [68,69]. In seeking to extend COVID-19 mandate exemption policies to childhood vaccines, Republican legislators are signifying their willingness to capitalize on COVID-19 conflict to weaken the most important governance instrument currently in use to ensure the country’s children are vaccinated. Such action is not limited to legislative actors: in late April of 2023 a Mississippi judge created a religious exemption by fiat, reversing a 1970 court decision that removed parents’ right to refuse vaccines on religious grounds [25].

Future research should track attempts to change mandatory childhood vaccination policies in the United States, including attempts to remove NMEs, but also attempts to introduce more accessible exemptions or even remove mandates altogether. Such research should track the success of bills and record the political orientation of those actors advancing them. Some efforts have been undertaken along these lines for COVID-19 vaccine mandates [70]. It will be important for that work to continue, for it to include policy activity focused on school vaccine mandates, and for it to pay attention to the unique risks and specific discourses that surround religious exemptions in particular.

It is also important for researchers to undertake detailed evaluation of the processes through which changes to exemption policies are introduced and how they fare in implementation, paying attention to any unintended consequences. Excellent work undertaken in California, in particular, can provide a blueprint for deep dives into other jurisdictions, and demonstrates the importance of employing various methodologies, including qualitative methodologies, in order to triangulate findings and uncover the impacts of specific processes [5,6,8,9,71]. Research undertaken in Vermont demonstrates that these deep dives can form part of larger, national studies [57]. A recently-developed guide for considering vaccine mandates at various stages of the policy cycle may also prove useful [72].

American governments have largely exhausted the ‘West Coast Period’ of behavioral-science-informed tweaks to vaccine mandates. This compromise position is no longer tenable. In the near future, few states are likely to attempt to modify their NME policies to make waivers more difficult to receive. Instead, policy reforms will likely focus on nationalizing NME abolition (supported by Democrats, the American Medical Association, and the American Academy of Pediatrics). By contrast, political and judicial actors in ‘red’ states will continue to try and make it easier for families to avoid vaccinating their children [70]. The ‘California model’ is unlikely to succeed nationally.

The impact of COVID-19 political polarization about masks, vaccines, mandates, and other state emergency measures has accelerated threats to existing childhood vaccine mandates. These conflicts have drawn the participation of new actors, including political opportunists, and have facilitated the spread of misinformation and disinformation about a range of health measures and policies. All of this is bad news for childhood vaccination policy. Along these lines, scholars from the Lancet Commission on Vaccine Refusal, Acceptance, and Demand in the U.S.A suggest that ‘the pandemic might be an inflection point in the acceptability and sustainability of school vaccine mandates in the U.S.A’ and foresee the possibility of ‘a new fervor for challenging school mandates’ [73,74]. Accordingly, we should anticipate a more symmetric political battle over NMEs in the coming years: Democrats are likely to continue to fight for the tightening of school vaccine mandates (primarily through eliminating NMEs), while Republicans are likely to make complementary moves toward weakening or eliminating school vaccine mandates (including by making NMEs more permissive). We are witnessing the dawn of a new period in which political opportunists and activists – hardened and ‘disinformed’ through COVID-19 political conflicts – use arguments about freedom and religious liberty to attack and weaken existing childhood vaccine mandates.

8. Notes

Interviews conducted with the author:

Catherine Flores Martin, Executive Director of the California Immunisation Coalition, 13 June 2019

Kris Calvin, CEO of AAP California, 1 July 2019

Dr Richard Pan, California State Senator, 14 June 2019

Leah Russin, Advocate, Vaccinate California, 11 July 2019

Article highlights

• American state governments are removing long-standing nonmedical exemptions to vaccine mandates for school entry.

• Advocates of removing exemptions may believe that they are re-instituting ‘pure’ mandate policies.

• Removing nonmedical exemptions transforms mandates from their original function, which was to motivate parents to vaccinate.

• Exemption removal generates challenges for implementation and may promote further post-COVID political polarization about vaccination policy.

• American states should take lessons from other jurisdictions about how to implement coercive vaccine policy effectively.

Declaration of interest

K Attwell is a specialist advisor to the Australian Technical Advisory Group on Immunisation. She is a past recipient of a Discovery Early Career Researcher Award funded by the Australian Research Council of the Australian Government (DE19000158). She leads the ‘Coronavax’ project, which is funded by the Government of Western Australia. She leads ‘MandEval: Effectiveness and Consequences of Australia’s COVID-19 Vaccine Mandates,’ a new project funded by the Medical Research Future Fund of the Australian Government (2019107). All funds were paid to her institution. Funders are not involved in the conceptualization, design, data collection, analysis, decision to publish, or preparation of manuscripts. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

K Attwell and M Navin conceptualized and designed the study, drafted the initial manuscript, and critically reviewed and revised the manuscript. K Attwell conducted the interviews. Both authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Acknowledgments

The authors thank the anonymous reviewers of an earlier draft of the manuscript for their helpful suggestions. They also thank their wider network of supporters and research collaborators, and the interviewees who gave their time and thoughts to this research project.

Additional information

Funding

This manuscript was funded by the Australian Research Council of the Australian Government, under DE190100158. The Australian Research Council had no role in this manuscript or the research underpinning it.