US state vaccine mandates did not influence COVID-19 vaccination rates but reduced uptake of COVID-19 boosters and flu vaccines compared to bans on vaccine restrictions

Edited by Mary Waters, Harvard University, Cambridge, MA; received August 9, 2023; accepted December 11, 2023
February 15, 2024
121 (8) e2313610121

Abstract

During the COVID-19 pandemic, some US states mandated vaccination for certain citizens. We used state-level data from the CDC to test whether vaccine mandates predicted changes in COVID-19 vaccine uptake, as well as related voluntary behaviors involving COVID-19 boosters and seasonal influenza vaccines. Results showed that COVID-19 vaccine adoption did not significantly change in the weeks before and after states implemented vaccine mandates, suggesting that mandates did not directly impact COVID-19 vaccination. Compared to states that banned vaccine restrictions, however, states with mandates had lower levels of COVID-19 booster adoption as well as adult and child flu vaccination, especially when residents initially were less likely to vaccinate for COVID-19. This research supports the notion that governmental restrictions in the form of vaccination mandates can have unintended negative consequences, not necessarily by reducing uptake of the mandated vaccine, but by reducing adoption of other voluntary vaccines.
For purposes of public health, governments institute behavioral restrictions to gain compliance from citizens. During the COVID-19 pandemic, some US states mandated vaccination among targeted groups (e.g., state employees) to increase vaccine uptake. However, do such restrictions lead citizens to act (counter)productively? The theory of psychological reactance serves as one longstanding explanation for why freedom restrictions in the form of governmental mandates cause people to reject the advocated behavior or otherwise have unintended consequences (1). Research examining COVID-19 vaccine mandates, however, has been mixed. Although lab-based experimental studies show that mandates can reduce vaccination intentions relative to voluntary policies (24), other muti-experiment projects (5) indicate increased intentions stemming from mandates. Surveys associate the motivational state of psychological reactance in response to COVID-19 mandates with unvaccinated status (6) and reduced intentions to vaccinate for COVID-19 (7) but also find that reactance is not associated with COVID-19 vaccination behavior (8).
We report the results of a natural experiment examining the impact of state-level COVID-19 mandates on vaccination behavior using population-level epidemiological data from the US Centers for Disease Control and Prevention (CDC). We examine changes in COVID-19 vaccination rates before and after the imposition of state-level mandates in order to extend existing experimental and survey research by connecting state-level vaccination requirements directly with vaccination behavior of the general public. We also investigate the adoption of subsequent voluntary vaccines (i.e., COVID boosters and seasonal flu) as ways citizens restore freedoms threatened by COVID-19 vaccine mandates to generate insights about the unintended consequences of state-level mandates. Finally, this project makes it possible to compare the consequences of legislation aimed at protecting freedom. That is, we examine the implications of state-level bans preventing the imposition of COVID-19 vaccine requirements and thereby preserving the public’s autonomy over vaccination. Detailed descriptions of all measures and analyses can be found in SI Appendix; data and script to reproduce the analyses are available at https://osf.io/j3tqd/.

Results

We first examined the effects of state-level COVID-19 vaccine mandates on COVID-19 vaccine adoption. Among the 19 states that imposed a COVID-19 vaccine mandate, we used a difference-in-differences test (9) to determine whether change in weekly vaccination rates per 100k residents during the 8 wk before and after a vaccine mandate varied based on state-level baseline attitudes toward vaccine mandates. Baseline attitudes were measured at the state level just before the vaccine was available to the public as part of the COVID States Project. They were included in our analysis because research suggests that the efficacy of vaccine mandates depends on how people feel about the behavior before restrictions are instituted (4, 10). We also covaried the proportion of people already vaccinated in a state prior to our 8-wk sampling frame (Table 1). There was no statistically significant difference in weekly COVID-19 vaccination rates before and after the imposition of a mandate (b = −42.27, 95% CI [−105.40, 20.86]), and the difference in vaccination rates from before to after mandates did not depend on baseline attitudes toward COVID-19 vaccine mandates (i.e., mandate imposition × baseline attitude; b = −153.05, 95% CI [−1210.10, 904.00]). A post hoc power analysis showed that power was sufficient to detect moderate and large effects but not small effects. As a robustness test, we re-conducted the analyses varying the sampling window before and after mandates (i.e., 4 and 16 wk before/after) and observed the same trends. In sum, the data we examined from the CDC indicated that the imposition of mandates to require COVID-19 vaccination was not associated with changes in COVID-19 vaccine adoption rates.
Table. 1.
Change in new COVID-19 vaccination rates weekly based on COVID-19 mandate imposition and baseline attitudes
 New weekly vaccinations per 100kNew weekly vaccinations per 100kNew weekly vaccinations per 100k
 b95% CIpb95% CIpb95% CIp
Intercept1567.381314.87 to 1819.90<0.0011150.50807.62 to 1493.38<0.0011105.48642.19 to 1568.76<0.001
Proportion of state vaccinated−1667.37−2100.01 to −1234.72<0.001−1802.00−2310.83 to − 1293.18<0.001−1802.89−2312.56 to −1293.22<0.001
Baseline attitude toward vaccine mandate
   882.20304.17 to 1460.220.003959.91170.46 to 1749.360.017
Mandate (0 = before; 1 = after)   −42.27−105.40 to 20.860.18947.32−574.70 to 669.350.881
Baseline attitude × mandate      −153.05−1210.10 to 904.000.776
Observations297297297
R2/R2 adjusted0.163/0.1600.198/0.1900.198/0.187
Notes. The outcome variable was new weekly COVID-19 vaccinations per 100k residents. 95% CI = 95% confidence interval. Bold values indicate p < .05.
Our analyses focusing on COVID-19 booster and flu vaccine adoption, however, offered evidence that COVID-19 vaccine mandates had unintended consequences. Using data from the CDC, a series of multilevel models were tested to examine state-level differences in weekly COVID-19 booster adoption between November 2021 (when boosters became available) and May 2022 in the 19 states with a COVID-19 vaccine mandate and 22 states that enacted legislation to ban COVID-19 vaccination requirements. States with bans offer a unique comparison group because bans prevented the imposition of requirements and thereby functioned to protect individuals’ freedom over the decision to receive a vaccination. Our outcome variable was the proportion of citizens in a state who received a COVID-19 booster among those who were eligible by virtue of having been vaccinated 6 or more months prior. We further included state-level vaccination rates (i.e., initial COVID-19 vaccination) and its interaction with legislation type (i.e., vaccine mandate vs. ban) in our model. Vaccination rate was included to account for the general inclination toward COVID-19 vaccination in a state; because this analysis was limited to the subpopulation of people in a state who were eligible for a booster, the state-level baseline attitude measure was no longer appropriate. A random intercept was included for each state. The results indicated that the proportion of eligible people who received a COVID-19 booster was smaller in states that mandated vaccination than in states that banned vaccine requirements (b = −0.07, 95% CI [−0.13, −0.02]). Additionally, the interaction between legislation type and the proportion of a state who received a COVID-19 vaccine was statistically significant (b = 0.77, 95% CI [0.55, 0.98]). Although COVID-19 booster uptake was lower in states that imposed mandates compared to bans of vaccination requirements, the difference was larger among states that had lower vaccination levels (Fig. 1A). This pattern is consistent with a tendency to respond negatively to COVID-19 vaccine mandates relative to bans, especially among states whose residents initially were less inclined toward getting vaccinated.
Fig. 1.
(AC) report COVID-19 booster (A) and flu vaccination [adults (B); children (C)] adoption in states that imposed a COVID-19 vaccination mandate compared to states that banned vaccination requirements. (A and C) report the interaction between COVID-19 vaccination rates and state vaccination legislation (ban vs. mandate.)
We further examined two datasets from the CDC reporting state-level seasonal flu vaccination among adults and children during the 2021–2022 flu season spanning the same general time period as the booster analysis and using the same data analysis procedures. Controlling for the proportion of people in a state who received a COVID-19 vaccine, adults in states with mandates were less likely to report receiving the seasonal flu vaccine than adults in states where COVID-19 vaccination requirements were banned (b = −0.12, 95% CI [−0.16, −0.08]; see Fig. 1B). This finding replicated among children (b = −0.18, 95% CI [−0.23, −0.12]), where an interaction between state legislation type and COVID-19 vaccination rates was also observed (b = 0.15, 95% CI [0.04, 0.27]). The results for the interaction followed the same pattern as for COVID-19 boosters (Fig. 1C). Although flu vaccination among children was generally lower in states that imposed mandates compared to bans, the difference was larger among states that had lower levels of COVID-19 vaccination. The results from the flu analysis collectively offer further evidence of unintended consequences in states that imposed COVID-19 vaccination mandates compared to bans on vaccination requirements.

Discussion

Our results suggest two primary conclusions about the influence of US legislation mandating COVID-19 vaccination and legislation formally banning COVID-19 vaccine requirements. One is that mandates did not significantly influence residents’ COVID-19 vaccine uptake. Although readers should note that we only had sufficient power to detect medium to large effects, the results offer evidence that mandates did not substantially impact targeted vaccination behavior. A second key conclusion is that relative to state legislation banning vaccine requirements, state COVID-19 vaccination mandates had unintended consequences involving reduced adoption of COVID-19 boosters and seasonal flu vaccines, especially when initial COVID-19 vaccination was low. The results from this project help to clarify inconsistencies and extend prior research on COVID-19 and vaccination mandates more generally. Despite the freedom threat posed by vaccine mandates, our findings complement prior research by using population-level epidemiological data to show that mandates do not reduce adoption of the required vaccine (8) but of other voluntary ones (24).
More broadly, the results underscore the challenges of promoting public health through vaccination. That the imposition of state COVID-19 mandates was not associated with a reduction in vaccination may stem from a reasonable desire to retain employment or avoid social sanctions despite limitations to freedom. The data from this project indicate that people responded to the limitation to their freedom that stems from vaccination mandates by being less likely to adopt COVID-19 boosters and influenza vaccines relative to states banning vaccination requirements. In this way, legislation designed to protect the public’s freedom to choose whether to be vaccinated for COVID-19 appears to have been more productive in encouraging related vaccinations. These findings provide evidence to support the concerns of scholars and practitioners that selective vaccination mandates can have harmful—or at least unintended—consequences for public health, which need consideration when creating such policies (11, 12).
Three caveats deserve note in considering our project. First, as explained in SI Appendix, mandate states had higher absolute booster adoption rates in terms of the total number of people boosted compared to states that enacted bans on vaccine requirements. Our analyses reported here focused on the proportion of eligible people (i.e., were vaccinated 6 mo prior) who received a booster. Second, readers should note that this project deviates from prior research on vaccine mandates by virtue of examining this phenomenon at the group (i.e., state) level rather than among individuals. Third, in evaluating the imposition of mandates and bans during the pandemic, it is possible that the various groups considered in this project differed in ways that were not measured.

Materials and Methods

Data for this project were aggregated from several sources, including the US CDC, COVID States Project, Ballotpedia, and the US Census. We identified states that imposed a mandate requiring vaccination or a ban preventing the imposition of vaccine requirements. We then constructed measures of state-level baseline vaccine mandate attitudes, COVID-19 vaccination rates, COVID-19 booster rates, and flu vaccination rates. Detailed descriptions of all measures and analyses can be found in SI Appendix; data and script to reproduce the analyses are available at: https://osf.io/j3tqd/.

Data, Materials, and Software Availability

Anonymized Data were aggregated from multiple existing sources. The data and script necessary to replicate our analyses are available at the OSF have been deposited in Open Science Framework (https://osf.io/j3tqd/) (13). Previously published data were used for this work (1417).

Acknowledgments

There is no funding source to report.

Author contributions

S.A.R. and A.S.R. designed research; S.A.R. analyzed data; and S.A.R. and A.S.R. wrote the paper.

Competing interests

The authors declare no competing interest.

Supporting Information

Appendix 01 (PDF)

References

1
J. W. Brehm, A Theory of Psychological Reactance (Academic Press, 1966).
2
P. Sprengholz, C. Betsch, Zero-sum or worse? Considering detrimental effects of selective mandates on voluntary childhood vaccinations. J. Pediatr. 240, 318–319 (2022).
3
P. Sprengholz, C. Betsch, R. Böhm, Reactance revisited: Consequences of mandatory and scarce vaccination in the case of COVID-19. Appl. Psychol. Health Well-Being 13, 986–995 (2021).
4
P. Sprengholz, L. Felgendreff, R. Böhm, C. Betsch, Vaccination policy reactance: Predictors, consequences, and countermeasures. J. Health Psychol. 27, 1394–1407 (2022).
5
D. Albarracin, H. Jung, W. Song, A. Tan, J. Fishman, Rather than induing psychological reactance, requiring vaccination strengthens intentions to vaccinate in US populations. Sci. Rep. 11, 20796 (2021).
6
P. Sprengholz, L. Henkel, R. Böhm, C. Betsch, Different interventions for COVID-19 primary and booster vaccination? Effects of psychological factors and health policies on vaccine uptake. Med. Dec. Mak. 43, 239–251 (2023).
7
A. de Figueiredo, H. J. Larson, S. D. Reicher, The potential impact of vaccine passports on inclination to accept COVID-19 vaccinations in the United Kingdom: Evidence from a large cross-sectional survey and modeling study. EClinicalMedicine 40, 101109 (2021).
8
I. A. M. Verpaalen et al., Psychological reactance and vaccine uptake: A longitudinal study. Psychol. Health, https://doi.org/10.1080/08870446.2023.2190761 (2023).
9
J. B. Dimick, A. M. Ryan, Methods for evaluating changes in health care policy: The difference-in-differences approach. JAMA 22, 2401–2402 (2014).
10
R. A. Wicklund, Freedom and Reactance (Lawrence Erlbaum, 1974).
11
K. Bardosh et al., The unintended consequences of COVID-19 vaccine policy: Why mandates, passports and restrictions may cause more harm than good. BMJ Global Health 7, e008684 (2022).
12
S. B. Omer, C. Betsch, J. Leask, Mandate vaccination with care. Nature 571, 469–472 (2019).
13
S. A. Rains, A. S. Richards, State COVID-19 Vaccination Mandatesand Vaccine Adoption Dataset. Open Science Framework. https://osf.io/j3tqd/. Deposited 3 August 2023.
14
CDC, COVID-19 vaccinations in the United States [Dataset] (2022), https://data.cdc.gov/Vaccinations/COVID-19-Vaccinations-in-the-United-States-Jurisdi/unsk-b7fc. Accessed 16 June 2022.
15
CDC, Cumulative influenza vaccination coverage by race/ethnicity and age group, NIS-ACM [Dataset] (2023), https://data.cdc.gov/Vaccinations/Cumulative-Influenza-Vaccination-Coverage-by-Race-/8dyx-9z99. Accessed 6 January 2023.
16
CDC, Weekly cumulative influenza vaccination coverage, children 6 months through 17 years, United States [Dataset] (2023), https://data.cdc.gov/Vaccinations/Weekly-Cumulative-Influenza-Vaccination-Coverage-C/eudc-n39h. Accessed 6 May 2023.
17
D. Lazer et al., The COVID States Project [Dataset] (2022), https://lazerlab.shinyapps.io/Behaviors_During_COVID/. Accessed 12 July 2022.

Information & Authors

Information

Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 121 | No. 8
February 20, 2024
PubMed: 38359292

Classifications

Data, Materials, and Software Availability

Anonymized Data were aggregated from multiple existing sources. The data and script necessary to replicate our analyses are available at the OSF have been deposited in Open Science Framework (https://osf.io/j3tqd/) (13). Previously published data were used for this work (1417).

Submission history

Received: August 9, 2023
Accepted: December 11, 2023
Published online: February 15, 2024
Published in issue: February 20, 2024

Keywords

  1. vaccine mandates
  2. government restriction
  3. personal freedom
  4. social influence

Acknowledgments

There is no funding source to report.
Author Contributions
S.A.R. and A.S.R. designed research; S.A.R. analyzed data; and S.A.R. and A.S.R. wrote the paper.
Competing Interests
The authors declare no competing interest.

Authors

Affiliations

Department of Communication, University of Arizona, Tucson, AZ 85721-0025
Department of Communication Studies, Furman University, Greenville, SC 29613-1000

Notes

1
To whom correspondence may be addressed. Email: [email protected].

Metrics & Citations

Metrics

Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.


Citation statements




Altmetrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

View Options

View options

PDF format

Download this article as a PDF file

DOWNLOAD PDF

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Personal login Institutional Login

Recommend to a librarian

Recommend PNAS to a Librarian

Purchase options

Purchase this article to get full access to it.

Single Article Purchase

US state vaccine mandates did not influence COVID-19 vaccination rates but reduced uptake of COVID-19 boosters and flu vaccines compared to bans on vaccine restrictions
Proceedings of the National Academy of Sciences
  • Vol. 121
  • No. 8

Media

Figures

Tables

Other

Share

Share

Share article link

Share on social media