School Mandate and Influenza Vaccine Uptake Among Prekindergartners in New York City, 2012–2019

Abstract

New York City (NYC) introduced a universal prekindergarten program in 2014 that mandated influenza vaccination for enrollment. We conducted a difference-in-difference-in-differences study to evaluate the program using 2012 to 2019 MarketScan claims data. After the introduction of the program, influenza vaccine uptake among four-year-old children in NYC during the subsequent seasons increased by 6.3 to 9.8 percentage points compared with the rest of New York State. (Am J Public Health. 2022;112(5):719–723. https://doi.org/10.2105/AJPH.2022.306765)

School-entry vaccination mandates have been implemented in the United States to reduce the risk of vaccine-preventable disease outbreaks by increasing childhood vaccination coverage. Under such mandates, vaccination records are required for school entry, and students must be up-to-date on required vaccinations, with exemptions allowed under certain circumstances. Studies have shown that school-entry vaccination mandates effectively increase coverage with various vaccines,¹ but there is limited evidence on influenza vaccination.^2,3 One possible reason for this is lack of widely implemented school-entry mandates: until July 2020, there were only five states (Connecticut, New Jersey, Ohio, Pennsylvania, Rhode Island) and one city (New York City) mandating annual influenza vaccination for school entry, all at child care or preschool levels,⁴ probably because of political or legal backlash from parents who view influenza vaccination as optional. Influenza vaccination coverage among children aged six months to 17 years in the United States remains below the Healthy People 2030 target of 70%.^5,6 Rigorous empirical evidence on the associations between school-entry mandates and influenza vaccination could be helpful for policymakers seeking to design strategies to increase children’s influenza vaccination uptake.⁷

INTERVENTION

A universal prekindergarten (UPK) program in New York City (NYC) provided an opportunity to fill this knowledge gap. In the fall of 2014, NYC introduced Pre-K for All, a program that provided one year of free prekindergarten (Pre-K) education for approximately 53 000 four-year-old children born in 2010. Children residing in NYC were eligible for UPK starting in September of the year when they turned four years old. The program became universal in the fall of 2015; a seat was guaranteed for any eligible child who applied for enrollment. Since then, approximately 70 000 children have been enrolled yearly, accounting for about 60% of the four-year-old population in NYC. The program has built-in health policies, which has improved the health of children from low-income families.⁸ Specifically, NYC’s UPK program requires a series of childhood vaccinations for enrollment, consistent with the requirement for Pre-K enrollment in New York State (NYS).^9,10 Additionally, the NYC program requires one dose of influenza vaccine during the current influenza season by December 31 of the Pre-K school year, whereas NYS does not have such a requirement for Pre-K enrollment.^9,10 A few exemptions are allowed in both NYC and NYS.⁷ A similar influenza vaccine mandate was introduced for city-regulated child care programs in 2014 without citywide program expansion.⁷ In total, the mandates affected about 150 000 children aged six to 59 months.

PLACE AND TIME

The program was implemented in NYC, which had an estimated population of 8 269 194 in 2019, about 7.4% of which (613 571) comprised children aged younger than six years. In 2019, the rest of NYS had an estimated population of 10 943 619, 6.6% of which (718 657) comprised children aged younger than six years.

The mandate was introduced with the UPK program in September 2014, was suspended in December 2015, and has been in force since being reinstated in June 2018. The mandate for city-regulated child care programs experienced the same suspension and reinstatement. We analyzed data from August 1, 2012 through May 31, 2019.

PERSON

The treatment group consisted of Pre-K--eligible children aged four years residing in NYC. The control group consisted of Pre-K--ineligible children residing in NYC (three or five years of age) or the rest of NYS (three through five years of age). We used the MarketScan Commercial Claims and Encounters databases that included children who were privately insured. Each year from 2012 to 2019, the percentage of children aged younger than six years with private health insurance was 46% to 51% in NYC and 63% to 65% in the rest of NYS.

PURPOSE

The mandate was imposed to protect prekindergartners from getting and spreading influenza, especially to older or more medically vulnerable household members, and potentially to reduce community influenza spread.

IMPLEMENTATION

The mandate was enacted in 2014 by the NYC Board of Health and the NYC Department of Health and Mental Hygiene. In mid-December 2015, the mandate for influenza vaccination was suspended by the NYS lower courts in response to a lawsuit against the mandate; in June 2018, the highest court of NYS upheld and reinstated the mandate.⁷

EVALUATION

Using the 2012–2019 MarketScan Commercial Claims and Encounters databases,¹¹ we used a difference-in-difference-in-differences (DDD) approach to evaluate whether the UPK program was associated with changes in children’s influenza vaccination uptake (Appendix Figure A, available as a supplement to the online version of this article at http://www.ajph.org). The data and methods are described in detail in the Appendix (Data and Measures, Methods).

Our study sample consisted of 279 941 observations of 178 873 children. Appendix Table A summarizes children’s characteristics that were controlled in DDD analyses, by influenza season and birth cohort. Despite some statistical significance, overall, the differences in children’s characteristics by age were small.

From the 2012–2013 to the 2018–2019 influenza season, influenza vaccination coverage in NYC increased from 53.1% to 68.5% among three-year-old children, from 50.4% to 69.2% among four-year-old children, and from 47.2% to 55.3% among five-year-old children. During the same period, in the rest of NYS, influenza vaccination coverage increased from 56.8% to 60.1% among three-year-old children, from 54.5% to 55.6% among four-year-old children, and from 51.6% to 53.2% among five-year-old children (Table 1). The substantial increases in coverage among three- and four-year-old children in NYC were likely due to the influenza vaccination mandates introduced to child care and preschools.

TABLE 1—

Influenza Vaccination Coverage Among Children Aged 3–5 Years: New York City (NYC) and the Rest of New York State (NYS), 2012–2013 to 2018–2019 Influenza Seasons

	NYC Children’s Age, Years			Rest of NYS Children’s Age, Years
3	4	5	3	4	5
2012–2013 influenza season
No. of children	6 861	6 815	7 032	6 863	7 232	7 884
Birth year	2009	2008	2007	2009	2008	2007
% vaccinated (95% CI)	54.1 (52.9, 55.3)	51.3 (50.1, 52.5)	48.0 (46.8, 49.1)	56.8 (55.7, 58.0)	53.5 (52.3, 54.6)	51.6 (50.5, 52.7)
2013–2014 influenza season
No. of children	9 555	9 483	9 790	12 733	13 067	13 788
Birth year	2010	2009	2008	2010	2009	2008
% vaccinated (95% CI)	52.1 (51.1, 53.1)	49.4 (48.4, 50.4)	46.3 (45.3, 47.3)	56.7 (55.9, 57.8)	55.5 (54.6, 56.3)	51.6 (50.7, 52.4)
2014–2015 influenza season
No. of children	5 742	6 088	6 142	9 927	10 465	10 660
Birth year	2011	2010	2009	2011	2010	2009
% vaccinated (95% CI)	57.1 (55.8, 58.4)	58.0 (56.7, 59.2)	48.0 (46.7, 49.2)	54.6 (53.6, 55.5)	52.8 (51.9, 53.8)	50.0 (49.0, 50.9)
2015–2016 influenza season
No. of children	5 630	5 810	6 166	8 458	8 833	9 373
Birth year	2012	2011	2010	2012	2011	2010
% vaccinated (95% CI)	56.6 (55.3, 57.9)	59.7 (58.4, 61.0)	48.8 (47.6, 50.0)	53.2 (52.2, 54.3)	49.8 (48.7, 50.8)	46.1 (45.1, 47.1)
2016–2017 influenza season
No. of children	4 229	4 032	4 240	3 072	3 085	3 221
Birth year	2013	2012	2011	2013	2012	2011
% vaccinated (95% CI)	47.6 (46.0, 49.1)	44.0 (42.4, 45.5)	43.8 (42.3, 45.3)	53.1 (51.3, 54.9)	51.8 (50.1, 53.6)	47.7 (46.0, 49.4)
2017–2018 influenza season
No. of children	4 163	4 370	4 285	3 162	3 267	3 376
Birth year	2014	2013	2012	2014	2013	2012
% vaccinated (95% CI)	57.7 (56.1, 59.2)	54.8 (53.3, 56.3)	50.3 (48.8, 51.8)	56.1 (54.4, 57.8)	55.3 (53.6, 57.0)	51.2 (49.5, 52.9)
2018–2019 influenza season
No. of children	5 798	5 910	6 316	4 218	4 329	4 471
Birth year	2015	2014	2013	2015	2014	2013
% vaccinated (95% CI)	68.5 (67.3, 69.7)	69.2 (68.1, 70.4)	55.3 (54.1, 56.5)	60.1 (58.6, 61.5)	55.6 (54.1, 57.0)	53.2 (51.7, 54.6)

Note. CI = confidence interval.

Source. MarketScan Commercial Claims and Encounters Database.¹¹

Table 2 shows the estimated changes (in percentage points) in influenza vaccination uptake that were associated with the UPK program. After the introduction of UPK influenza vaccination mandate, uptake by four-year-old children in NYC increased by 6.3 percentage points (95% confidence interval [CI] = 2.5, 12.0) during the 2014–2015 influenza season. The increase mostly occurred by the end of December, when the uptake increased by 5.5 percentage points (95% CI = 1.5, 13.3). The cumulative instantaneous probability of influenza vaccination further confirmed that the increases mainly occurred from mid-November to mid-January (Appendix Figure B).

TABLE 2—

Change in Influenza Vaccination Uptake Among Children Aged 4 Years Relative to Children Aged 3 and 5 Years in New York City Compared With Rest of New York State: 2012–2013 to 2018–2019 Influenza Seasons

	No. of Children	Percentage Point Change (95% CI) in Uptake During Influenza Season (August 1–May 31)
Main analysis (2014–2015 influenza season)	117 440	6.3*,a (2.5, 12.0)
Additional analyses
Placebo (2013–2014 influenza season)	111 103	−2.2 (–7.2, 5.0)
2015–2016 influenza season	112 686	8.3 (5.4, 15.5)
2016–2017 influenza season	90 295	−1.9 (–4.9, 2.8)
2017–2018 influenza season	91 039	0.5 (–6.9, 5.9)
2018–2019 influenza season	99 458	9.8 (7.5, 15.9)

Note. CI = confidence interval. Each cell represents a separate regression with the period during which change in influenza vaccination uptake was estimated noted in the row headings. We calculated 95% confidence intervals based on standard errors that were clustered at the metropolitan statistical area level using the wild cluster bootstrap method.

Source. MarketScan Commercial Claims and Encounters Database.¹¹

^a The increase in influenza vaccination uptake mostly occurred from August 1 to December 31 (5.5 percentage points; 95% CI = 1.5, 13.3).

^*P < .05, derived from 2-sided t test. For each independent variable, the regression includes gender, whether living in a metropolitan statistical area, and insurance plan type.

We conducted several additional analyses (Table 2). First, we found no statistically significant association between a placebo UPK program in 2013 and influenza vaccination uptake during the 2013–2014 influenza season (–2.2 percentage points; 95% CI = –7.2, 5.0), validating our DDD strategy. Second, we found a stronger association after the full UPK rollout in the 2015–2016 influenza season (8.3 percentage points; 95% CI = 5.4, 15.5). Third, there was no statistically significant association during the suspension of the mandate (–1.9 percentage points in the 2016–2017 influenza season, 95% CI = –4.9, 2.8; 0.5 percentage points in the 2017–2018 influenza season, 95% CI = –6.9, 5.9). Fourth, we obtained a statistically significant association after the reinstatement during the 2018–2019 influenza season (9.8 percentage points; 95% CI = 7.5, 15.9). Finally, the estimated associations differed by the age of children in the control group, probably affected by the mandate introduced to city-regulated child care that might increase influenza vaccination among three-year-old children. Meanwhile, five-year-old children were mostly in kindergartens, for which there was no influenza vaccine mandate. Detailed results from the main DDD and additional analyses are presented in Appendix Tables B through D.

Limitations include the fact that our study included privately insured children only. The associations for other children might be different.

ADVERSE EFFECTS

The mandate faces legal challenges and was suspended for two and half influenza seasons. Side effects of influenza vaccination were not systematically collected.

SUSTAINABILITY

During our study period, because of the legal challenge, the influenza vaccination mandate was implemented for only two and half influenza seasons. Longer-term evidence is needed to understand the sustainability. Influenza vaccinations for children are available at their pediatrician’s or primary care doctor’s office. Starting in January 2018, prekindergartners in NYS can also receive influenza vaccinations from certified pharmacists.

PUBLIC HEALTH SIGNIFICANCE

The NYC’s UPK program successfully got more children vaccinated against influenza by mid-January, which is prior to the peak of influenza activity in the United States in most seasons.¹² Our findings demonstrate that school-entry influenza vaccination mandates are potentially effective strategies to increase children’s influenza vaccination uptake, particularly in a large urban environment with suboptimal coverage, and help protect other more vulnerable household and low-income community members. Mandating influenza vaccination together with other required vaccinations is feasible, although the effectiveness might be weaker when tracking systems of immunization histories have not been well established. Our evaluation provides useful information for policymakers about feasibility, benefits, and potential legal challenges, when they consider mandating influenza vaccination for school-aged children.