Mental Health Status of Rhode Island Middle School and High School Students Before Versus During the COVID-19 Pandemic

Abstract

Background

The COVID-19 pandemic negatively affected adolescent mental health due to school closures, isolation, family loss/hardships, and reduced health care access.

Methods

We compared adolescent mental health in Rhode Island before versus during the pandemic, separately among middle and high schoolers. This serial cross-sectional study used Youth Risk Behavior Survey data from 2019 and 2021 (N=7,403). Multivariable logistic regression models estimated the association between year and mental health status, adjusting for sociodemographics.

Results

Middle schoolers in 2021 had higher odds of ever seriously considering suicide (22.6% vs. 16.7%) and ever attempting suicide (9.3% vs. 6.1%) compared to 2019. Among high schoolers, those in 2021 had higher odds of experiencing persistent sadness/hopelessness in the past year (37.4% vs. 32.0%). However, high schoolers in 2019 and 2021 had similar odds of considering suicide in the past year, while those in 2021 had lower odds of having attempted suicide in the past year (8.5% vs. 14.6%).

Conclusion

The COVID-19 pandemic may have worsened multiple aspects of adolescent mental health in Rhode Island, particularly among middle schoolers.

Implications for school health policy, practice, and equity

Promoting school connectedness, creating supportive environments, and diversifying the mental health workforce may help overcome adverse pandemic effects.

With the onset of the COVID-19 pandemic in March 2020, adolescents experienced school closures, social isolation, family economic hardship, family loss or illness, and reduced access to health care.¹ Previous research suggests that the pandemic has negatively affected adolescent mental health. For example, symptoms of depression and anxiety increased among adolescents globally during the pandemic,^2,3 and the percentage of mental health-related emergency department visits among adolescents aged 12 to 17 years nationally increased by 31.0% from 2019 to 2020.⁴ Additionally, in a 2021 survey, 37.1% of high school students nationally had experienced poor mental health during the pandemic and, in the past year, 44.2% persistently felt sad/hopeless, 19.9% had seriously considered attempting suicide, and 9.0% had attempted suicide.⁵

Some subgroups reported poor mental health more often than others during the COVID-19 pandemic. In 2021, 26.3% of lesbian, gay, and bisexual students, and 16.5% of other or questioning students, attempted suicide compared to 5.2% of heterosexual students.⁵ Additionally, during the pandemic, Black students were more likely to lose a parent or caregiver compared to other students, American Indian/Alaska Native students more often faced challenges staying connected with friends as a result of limited internet access, and Latino students more often reported poor mental health and loneliness.⁶ However, the extent to which this represents a disproportionate impact of the pandemic (versus a persistent disparity that preceded the pandemic) is uncertain. Additionally, most studies of the impact of the pandemic on adolescent mental health have been among adolescents of all ages, though its impact may have differed for middle and high school students.

Previous research suggests that school connectedness, which is the belief among students that adults and peers at school care about their learning and them as individuals,⁷ is a protective factor for mental health.⁸ Nationally, high school students who felt close to people at school during the pandemic were less likely than those who did not to report feelings of persistent sadness/hopelessness (35.4% vs. 52.9%), having seriously considered attempting suicide (14.0% vs. 25.6%), and having attempted suicide (5.8% vs. 11.9%).⁵ However, the impact of the pandemic on temporal trends in school connectedness and non-parental support is unknown.

The objective of this study was to compare the mental health status of students in Rhode Island before and during the COVID-19 pandemic, separately among middle and high school students. Based on national trends, we hypothesized that the prevalence of adverse mental health outcomes among middle and high students in Rhode Island increased during the pandemic. As secondary objectives, we compared students’ comfort in seeking non-parental support before versus during the pandemic and considered whether the pandemic impacted students differently based on sexual and gender identity or race and ethnicity. We hypothesized that, among middle and high school students, the prevalence of having non-parental support decreased during the pandemic and that sexual/gender and racial/ethnic minority students experienced a disproportionate adverse impact of the pandemic on mental health.

METHODS

Study design, data, and sample

We used Rhode Island Youth Risk Behavior Survey (YRBS) data from the Rhode Island Department of Health (RIDOH) to conduct a serial cross-sectional study. The Rhode Island YRBS is a component of the national Youth Risk Behavior Surveillance System, which was designed to assess health behaviors and risks of public-school students in grades 6 through 12.⁹ We used data from the 2019 and 2021 YRBS cycles to compare the mental health status of adolescents in Rhode Island before versus during the COVID-19 pandemic. RIDOH conducts the YRBS statewide, with data collected anonymously from a random sample of public middle and high schools in the spring of odd years.¹⁰ Due to COVID-19-related delays, the 2021 survey was conducted in the fall.

Measures

Exposure.

The exposure of interest in this study was the COVID-19 pandemic. Students who completed the YRBS in 2019 (i.e., before the pandemic) were considered “unexposed.” Students who completed the YRBS in 2021 (i.e., during the pandemic) were considered “exposed.”

Outcomes.

We considered three primary mental health outcomes in this study: persistent feelings of sadness/hopelessness, seriously considered attempting suicide, and attempted suicide. We also considered comfort in seeking non-parental support as a secondary outcome. The survey questions, response options, and analytic coding for these measures are available in Table 1.

Table 1.

Survey Question and Analytic Coding for Mental Health Outcomes and Sociodemographic Measures, YRBS 2019 and 2021

Measures	Survey Question	Survey Response Options	Analytic Coding
Mental Health Outcomes
Persistent Feelings of Sadness or Hopelessness	Middle School: Not assessed. High School: During the past 12 months, did you ever feel so sad or hopeless almost every day for two weeks or more in a row that you stopped doing some usual activities?	Middle School: Not assessed. High School: Yes and no.	Middle School: Not assessed. High School: Yes and no.
Seriously Considered Attempting Suicide	Middle School: Have you ever seriously thought about killing yourself? High School: During the past 12 months, did you ever seriously consider attempting suicide?	Middle School: Yes and no. High School: Yes and no.	Middle School: Yes and no. High School: Yes and no.
Attempted Suicide	Middle School: Have you ever tried to kill yourself? High School: During the past 12 months, how many times did you actually attempt suicide?	Middle School: Yes and no. High School: 0 times, 1 time, 2 or 3, 4 or 5, and 6 or more times.	Middle School: Yes and no. High School: Never and one or more times.
Comfort in Seeking Non-parental Support	Middle School: Besides your parents, how many adults would you feel comfortable seeking help from if you had an important question affecting your life? High School: Is there at least one teacher or other adult in your school that you can talk to if you have a problem?	Middle School: 0 adults, 1 adult, 2 adults, 3 adults, 4 adults, and 5 or more adults. High School: Yes, no, and not sure.	Middle School: None and one or more adults. High School: Yes and no or not sure.
Sociodemographic Measures
Sex	Middle School: What is your sex? High School: What is your sex?	Middle School: Male and female. High School: Male and female.	Middle School: Male and female. High School: Male and female.
Age	Middle School: How old are you? High School: How old are you?	Middle School: 10 years old or younger, 11 years old, 12 years old, 13 years old, 14 years old, 15 years old, and 16 years old or older. High School: 12 years old or younger, 13 years old, 14 years old, 15 years old, 16 years old, 17 years old, and 18 years old or older.	Middle School: 11 years old or younger, 12 years old, 13 years old, and 14 years old or older. High School: 14 years old or younger, 15 years old, 16 years old, 17 years old, and 18 years old or older.
Grade	Middle School: In what grade are you? High School: In what grade are you?	Middle School: 6th grade, 7th grade, 8th grade, and Ungraded or other grade. High School: 9th grade, 10th grade, 11th grade, 12th grade, and Ungraded or other grade.	Middle School: 6th grade, 7th grade, and 8th grade. High School: 9th grade, 10th grade, 11th grade, and 12th grade.
Race and Ethnicity	Middle School: What is your race? and Are you Hispanic or Latino? High School: What is your race? and Are you Hispanic or Latino?	Middle School: American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian/Other Pacific Islander, White, Hispanic/Latino, multiple races - Hispanic, and multiple races - non-Hispanic. High School: American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian/Other Pacific Islander, White, Hispanic/Latino, multiple races - Hispanic, and multiple races - non-Hispanic.	Middle School: Hispanic or Latino (any race), non-Hispanic Asian, non-Nispanic Black, non-Hispanic White, and non-Hispanic other or multiple races. High School: Hispanic or Latino (any race), non-Hispanic Asian, non-Hispanic Black, non-Hispanic White, and non-Hispanic other or multiple races.
Racial and Ethnic Minority Identity	Middle School: What is your race? and Are you Hispanic or Latino? High School: What is your race? and Are you Hispanic or Latino?	Middle School: American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian/Other Pacific Islander, White, Hispanic/Latino, multiple races - Hispanic, and multiple races - non-Hispanic. High School: American Indian/Alaska Native, Asian, Black or African American, Native Hawaiian/Other Pacific Islander, White, Hispanic/Latino, multiple races - Hispanic, and multiple races - non-Hispanic.	Middle School: Yes and no (non-Hispanic White). High School: Yes and no (non-Hispanic White).
Gender Expression Differs From Sex	Middle School: A person's appearance, style, dress, or the way they walk or talk may affect how people describe them. How do you think other people at school would describe you? What is your sex?	Middle School: Very feminine, mostly feminine, somewhat feminine, equally feminine and masculine, somewhat masculine, mostly masculine, and very masculine. Male and female.	Middle School: Yes and noa
Sexual and Gender Identity	High School: Which of the following best describes you? Some people describe themselves as transgender when their sex at birth does not match the way they think or feel about their gender. Are you transgender?	High School: Heterosexual (straight), gay or lesbian, bisexual, and not sure (2019). Heterosexual (straight), gay or lesbian, bisexual, I describe my sexual identity some other way, I am not sure about my sexual identity (questioning), and I do not know what this question is asking (2021). No, I am not transgender, Yes, I am transgender, I am not sure if I am transgender, and I do not know what this question is asking.	High School: LGBTQ+ and straight and cisgender.

Abbreviations: LGBTQ+, Lesbian, Gay, Bisexual, Transgender, Queer or Questioning; YRBS, Youth Risk Behavior Survey.

^aParticipants reporting female sex were classified as “yes” (i.e., gender expression differs from sex) if they reported “very masculine,” “mostly masculine,” “somewhat masculine,” or “equally feminine and masculine” gender expression and “no” (i.e., gender expression does not differ from sex) if they reported “very feminine,” “mostly feminine,” or “somewhat feminine” gender expression. The same approach was used for participants reporting male sex.

Other measures.

Several sociodemographic measures were selected for consideration in the analysis, including sex, age, grade, race/ethnicity, gender identity (middle school students), and sexual/gender identity (high school students).¹¹ The survey questions, response options, and analytic coding for these measures are available in Table 1.

Of note, the simplified racial/ethnic minority identity measure was used in exploratory analyses of potential effect modification by race/ethnicity. The more detailed race/ethnicity measure was used in all other analyses. Additionally, for middle school students, we created a gender identity measure by determining whether their gender expression differed from their sex. For high school students, we created a sexual/gender identity measure using information on their sexual orientation and gender identity.

Data analysis

We managed and analyzed the data in SAS, version 9.4 (Cary, North Carolina). All analyses were conducted separately among middle and high school students. First, we summarized the sociodemographic characteristics of students in 2019 and 2021. Second, we estimated the association between year (2021 vs. 2019) and each mental health outcome using multivariable logistic regression, adjusting for sociodemographic characteristics. Each model excluded students with missing covariate or outcome information. Of note, age and grade capture similar information, so only one was included in the models. We selected age for inclusion because the 2019 and 2021 surveys were conducted at different times of the year, which was expected to lead to some differences in the age distributions. Third, we evaluated whether the association between year and each outcome differed among sexual/gender minority and racial/ethnic minority subgroups by separately including interaction terms for year and sexual/gender minority identity and racial/ethnic minority identity. All analyses were weighted to account for the complex sampling design and ensure that results were representative of all public middle and high school students in Rhode Island.

RESULTS

Overall, 3,637 middle school and 3,766 high school students participated in the 2019 or 2021 YRBS cycles in Rhode Island and were included in this study.

Middle school students

Sociodemographic characteristics.

Among middle school students overall, most were aged 12 (32.5%) or 13 (35.2%) years, 34.3% were in 8th grade, about half were male (50.3%) and non-Hispanic White (52.8%), and 25.7% reported a gender expression that differed from their sex (Table 2). In bivariate analyses, the sociodemographic characteristics of middle school students were similar in 2019 and 2021 (each p>0.6), with one exception. Middle school students in 2021 were younger than in 2019 (p<0.001).

Table 2.

Rhode Island Middle School Student Characteristics Overall and by Year, YRBS 2019 and 2021

	By Year
	Overall N=3,637		2019 N=1,571		2021 N=2,066		P-Valuea
	Unweighted N	Weighted %	Unweighted N	Weighted %	Unweighted N	Weighted %
Sociodemographics
Sex							0.682
Female	1,784	48.5%	777	48.4%	1,007	48.7%
Male	1,807	50.3%	788	51.2%	1,019	49.3%
Missing	46	1.2%	6	0.4%	40	2.0%
Age (Years)							<0.001
11 or Younger	660	17.5%	204	12.6%	456	22.8%
12	1,174	32.5%	509	32.0%	665	33.0%
13	1,299	35.2%	539	34.3%	760	36.2%
14 or Older	500	14.6%	318	21.0%	182	7.8%
Missing	<5	b	<5	b	<5	b
Grade							0.814
6th	1,131	31.4%	514	32.3%	617	30.5%
7th	1,216	33.7%	539	33.5%	677	34.0%
8th	1,271	34.3%	502	33.2%	769	35.4%
Missing	19	0.2%	16	1.0%	<5	b
Race/Ethnicity							0.921
Hispanic or Latino (Any Race)	1,112	26.0%	414	25.0%	698	27.1%
Non-Hispanic Asian	127	2.9%	41	2.3%	86	3.5%
Non-Hispanic Black	331	8.2%	145	8.1%	186	8.3%
Non-Hispanic White	1,676	52.8%	808	54.7%	868	50.8%
Non-Hispanic Other or Multiple Races	236	5.6%	98	5.7%	138	5.5%
Missing	155	4.4%	65	4.2%	90	4.8%
Gender Expression Differs From Sex							0.667
Yes	961	25.7%	413	26.2%	548	25.1%
No	2,361	65.5%	1,007	64.4%	1,354	66.7%
Missing	315	8.8%	151	9.4%	164	8.2%
Mental Health
Ever Seriously Considered Attempting Suicide							<0.001
Yes	743	19.5%	268	16.7%	475	22.6%
No	2,825	78.7%	1,285	82.2%	1,540	74.9%
Missing	69	1.8%	18	1.1%	51	2.5%
Ever Attempted Suicide							0.0004
Yes	293	7.6%	99	6.1%	194	9.3%
No	3,290	91.0%	1,456	93.0%	1,834	89.0%
Missing	54	1.4%	16	0.2%	38	1.8%
Had a Non-Parent Adult With Whom Could Discuss Problems							0.108
One or More Adults	2,864	79.7%	1,226	78.6%	1,638	81.0%
None	688	18.2%	307	19.0%	381	17.4%
Missing	85	2.0%	38	2.4%	47	1.7%

Abbreviations: YRBS, Youth Risk Behavior Survey.

^aRao-Scott chi-squared test comparing the weighted samples for 2019 and 2021.

^bCounts of 1–4 and associated percentages were suppressed, in line with RIDOH’s Small Numbers Policy

Mental health outcomes.

In 2019 and 2021 overall, 19.5% of middle school students reported having ever seriously considered attempting suicide, 7.6% reported having ever attempted suicide, and 79.7% reported having an adult other than a parent with whom they could discuss problems. In bivariate analyses, the percentage of middle school students reporting having ever seriously considered attempting suicide (22.6% vs. 16.7%, p<0.001) and ever attempted suicide (9.3% vs. 6.1%, p<0.001) was higher in 2021 than 2019. However, the percentage of students reporting having a non-parent adult with whom they could discuss problems was similar in 2019 and 2021 (p=0.11).

In multivariable analyses, compared to 2019, middle school students in 2021 had higher odds of having ever seriously considered attempting suicide (adjusted odds ratio [AOR]=1.57, 95% confidence interval [CI]=1.24, 1.98) and ever attempted suicide (AOR=1.67, 95% CI=1.25, 2.22) (Table 3). However, middle school students in 2021 also had higher odds of having a non-parent adult with whom they could discuss problems compared to those in 2019 (AOR=1.18, 95% CI=1.01, 1.38).

Table 3.

Unadjusted and Adjusted Odds Ratios Estimating the Association Between Year and Specific Mental Health Outcomes Among Rhode Island Middle School Students, YRBS 2019 and 2021

	Ever Seriously Considered Attempting Suicide N=3,175a		Ever Attempted Suicide N=3,185a		Had a Non-Parent Adult With Whom Could Discuss Problems N=3,151a
	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b
Year
2019 (Ref)	1.00	1.00	1.00	1.00	1.00	1.00
2021	1.49 (1.15, 1.93)c	1.57 (1.24, 1.98)c	1.60 (1.17, 2.17)c	1.67 (1.25, 2.22)c	1.13 (0.96, 1.32)	1.18 ( 1.01, 1.38)c
Sex
Female	2.47 (1.97, 3.09)c	2.55 (2.00, 3.23)c	1.84 (1.43, 2.37)c	1.82 (1.33, 2.48)c	1.04 (0.81, 1.34)	1.05 (0.82, 1.34)
Male (Ref)	1.00	1.00	1.00	1.00	1.00	1.00
Age (Years)
11 or Younger (Ref)	1.00	1.00	1.00	1.00	1.00	1.00
12	1.19 (0.84, 1.70)	1.27 (0.93, 1.73)	1.33 (0.82, 2.17)	1.38 (0.82, 2.32)	0.95 (0.69, 1.31)	0.99 (0.69, 1.44)
13	1.45 (0.96, 2.19)	1.54 (1.00, 2.37)c	1.62 (0.93, 2.92)	1.56 (0.89, 2.73)	0.81 (0.63, 1.02)	0.80 (0.60, 1.08)
14 or Older	1.48 (1.00, 2.19)c	1.65 (1.12, 2.44)c	1.63 (0.86, 3.08)	1.87 ( 0.94, 3.72)	0.91 (0.64, 1.28)	1.05 (0.76, 1.45)
Race and Ethnicity
Hispanic or Latino (Any Race)	1.58 (1.25, 2.00)c	1.51 (1.20, 1.90)c	1.79 (1.06, 3.01)c	1.59 (0.94, 2.68)	0.58 (0.47, 0.70)c	0.57 (0.46, 0.71)c
Non-Hispanic Asian	1.19 (0.75, 1.87)	1.20 (0.73, 1.99)	1.24 (0.55, 2.82)	1.17 (0.50, 2.71)	0.83 (0.40, 1.73)	0.99 (0.55, 1.77)
Non-Hispanic Black	1.83 (0.72, 1.95)	1.15 (0.70, 1.89)	1.70 (0.94, 3.06)	1.60 (0.91, 2.84)	0.51 (0.41, 0.64)c	0.52 (0.41, 0.65)c
Non-Hispanic White (Ref)	1.00	1.00	1.00	1.00	1.00	1.00
Non-Hispanic Other or Multiple Races	1.39 (0.76, 2.55)	1.36 (0.71, 2.62)	1.33 (0.60, 2.97)	1.21 (0.54, 2.74)	0.78 (0.50, 1.21)	0.79 (0.52, 1.18)
Gender Expression Differs From Sex
Yes	1.96 (1.60, 2.39)c	2.00 (1.67, 2.41)c	2.17 (1.70, 2.77)	2.25 (1.86, 2.73)c	0.79 (0.66, 0.94)c	0.81 (0.67, 0.99)c
No (Ref)	1.00	1.00	1.00	1.00	1.00	1.00

Abbreviations: YRBS, Youth Risk Behavior Survey.

^aEach model excludes students with any missing covariate or outcome information.

^bEach multivariable model includes year, sex, age, race and ethnicity, and gender expression differs from sex

^cStatistically Significant

Of note, in multivariable analyses, female middle school students had higher odds of having ever seriously considered attempting suicide (AOR=2.55, 95% CI=2.00, 3.23) and ever attempted suicide (AOR=1.82, 95% CI=1.33, 2.48) compared to males. Students who reported a gender expression that differed from their sex had higher odds of having ever seriously considered attempting suicide (AOR=2.00, 95% CI=2.70, 2.77) and ever attempted suicide (AOR=2.25, 95% CI=1.86, 2.73), as well as lower odds of having a non-parent adult with whom they could discuss problems (AOR=0.81, 95% CI=0.67, 0.99), compared to those whose gender expression was the same as their sex. The number of middle school students with and without each mental health outcome by year and sociodemographic characteristic is available in Table S1.

In exploratory analyses of potential effect modification, the association between year and each mental health outcome did not differ by race/ethnicity (each p>0.6) or gender expression (each p>0.2).

High school students

Sociodemographic characteristics.

Among high school students overall, most were aged 15 (24.0%) or 16 (24.6%) years, more than a quarter were in 9th grade (26.7%), about half were male (51.7%), more than half were non-Hispanic White (55.2%), and 20.9% identified as lesbian, gay, bisexual, transgender, queer or questioning (LGBTQ+) (Table 4). In bivariate analyses, the sociodemographic characteristics of high school students were similar in 2019 and 2021 (each p>0.3), with the exception of age and sexual/gender identity. Compared to high school students in 2019, students in 2021 were younger (p<0.001) and more likely to identify as LGBTQ+ (25.5% vs. 20.9%, p<0.001).

Table 4.

Rhode Island High School Student Characteristics Overall and by Year, YRBS 2019 and 2021

	By Year
	Overall N=3,766		2019 N=1,613		2021 N=2,153		P-Valuea
	Unweighted N	Weighted %	Unweighted N	Weighted %	Unweighted N	Weighted %
Sociodemographics
Sex							0.682
Female	1,800	47.5%	791	48.6%	1,009	46.3%
Male	1,935	51.7%	807	50.6%	1,128	52.9%
Missing	31	0.8%	15	0.8%	16	0.8%
Age (Years)							<0.001
14 or Younger	660	15.6%	213	10.0%	447	21.3%
15	1,000	24.0%	481	24.3%	519	23.7%
16	918	24.6%	395	24.6%	523	24.5%
17	806	23.3%	315	23.6%	491	23.0%
18 or Older	376	12.3%	207	17.3%	169	7.3%
Missing	6	0.2%	<5	b	<5	b
Grade							0.999
9th	1,134	26.7%	563	26.2%	571	27.2%
10th	950	24.6%	409	25.0%	541	24.2%
11th	847	23.8%	349	23.9%	498	23.7%
12th	779	23.5%	266	23.5%	513	23.5%
Missing	56	1.4%	26	1.3%	30	1.4%
Race and Ethnicity							0.360
Hispanic or Latino (Any Race)	1,348	26.8%	521	2.2%	827	2.6%
Non-Hispanic Asian	132	2.4%	55	8.3%	77	8.9%
Non-Hispanic Black	361	8.6%	116	56.7%	245	53.7%
Non-Hispanic White	1,624	55.2%	776	3.1%	848	0.5%
Non-Hispanic Other or Multiple Races	241	5.2%	98	24.8%	143	28.8%
Missing	60	1.8%	47	4.9%	13	5.5%
Sexual and Gender Identity							<0.001
LGBTQ+	809	20.9%	263	16.2%	546	25.5%
Straight and Cisgender	2,759	74.9%	1,452	82.0%	1,307	67.8%
Missing	198	5.3%	43	1.8%	155	6.7%
Mental Health in Past 12 Months
Experienced Persistent Feeling of Sadness or Hopelessness							0.025
Yes	1,352	34.7%	530	32.0%	822	37.4%
No	2,374	64.3%	1,066	67.0%	1,308	61.5%
Missing	40	1.0%	17	0.9%	23	1.1%
Seriously Considered Attempting Suicide							0.017
Yes	571	15.0%	208	13.1%	363	16.9%
No	3,132	83.6%	1,374	85.4%	1,758	81.8%
Missing	63	1.4%	31	1.4%	32	1.3%
Attempted Suicide							<0.001
One or more times	432	11.5%	248	14.6%	184	8.5%
Never	3,266	87.0%	1,356	85.0%	1,910	89.0%
Missing	68	1.4%	9	0.4%	59	2.5%
Had Adult at School With Whom Could Discuss Problems							0.001
Yes	2,120	59.5%	1,007	65.4%	1,113	53.6%
No or Not Sure	1,435	35.3%	543	30.8%	892	39.7%
Missing	211	5.2%	63	3.7%	148	6.6%

Abbreviations: LGBTQ+, Lesbian, Gay, Bisexual, Transgender, Queer or Questioning; YRBS, Youth Risk Behavior Survey.

^aRao-Scott chi-squared test comparing the weighted samples for 2019 and 2021.

^bCounts of 1–4 and associated percentages were suppressed, in line with RIDOH’s Small Numbers Policy.

Mental health outcomes.

In 2019 and 2021 overall, 34.7% of high school students reported experiencing persistent feelings of sadness/hopelessness, 15.0% reported seriously considering suicide, and 11.5% reported attempting suicide in the past year. Additionally, 59.5% of high school students reported having an adult at school with whom they could discuss problems. In bivariate analyses, the percentage of high school students reporting persistent feelings of sadness/hopelessness (37.4% vs. 32.0%, p=0.03) and having seriously considered attempting suicide (16.9% vs. 13.1%, p=0.02) in the past year was higher in 2021 than 2019. However, the percentage of students reporting that they had attempted suicide in the past year (8.5% vs. 14.6%, p<0.001) and had an adult at school with whom they could discuss problems (53.6% vs. 65.4%, p=0.001) was lower in 2021 than 2019.

In multivariable analyses, compared to 2019, high school students in 2021 had higher odds of having experienced persistent feelings of sadness/hopelessness in the past year (AOR=1.18, 95% CI=1.01, 1.37) (Table 5). However, students in 2019 and 2021 had similar odds of having seriously considered attempting suicide in the past year (AOR=1.09, 95% CI=0.86, 1.38). In contrast, students in 2021 had lower odds of having attempted suicide in the past year (AOR=0.45, 95% CI=0.38, 0.54) and having an adult at school with whom they could discuss problems (AOR=0.74, 95% CI=0.60, 0.91) than those in 2019.

Table 5.

Unadjusted and Adjusted Odds Ratios Estimating the Association Between Year and Specific Mental Health Outcomes Among Rhode Island High School Students, YRBS 2019 and 2021

	Experienced Persistent Feeling of Sadness or Hopelessness N=3,468a		Seriously Considered Attempting Suicide N=3,453a		Attempted Suicide N=3,440a		Had Adult at School With Whom Could Discuss Problems N=3,374a
	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b	Unadjusted OR (95% CI)	Adjusted OR (95% CI)b
Year
2019 (Ref)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
2021	1.27 (1.08, 1.50)c	1.18 (1.01, 1.37)c	1.34 (1.06, 1.70)c	1.09 (0.86, 1.38)c	0.55 (0.46, 0.67)c	0.45 (0.38, 0.54)c	0.64 (0.51, 0.80)c	0.74 (0.60, 0.91)c
Sex
Female	2.62 (2.26, 3.03)c	2.19 (1.88, 2.57)c	2.07 (1.80, 2.39)c	1.46 (1.28, 1.67)c	1.16 (0.96, 1.40)	0.92 (0.77, 1.10)	0.88 (0.79, 0.98)c	0.92 (0.81, 1.05)
Male (Ref)	1 .00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Age (Years)
14 or Younger (Ref)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
15	1.02 (0.81, 1.27)	1.16 (0.95, 1.42)	0.85 (0.60, 1.19)	0.98 (0.70, 1.38)	1.01 (0.77, 1.32)	0.94 (0.66, 1.34)	1.18 (0.88, 1.58)	1.07 (0.78, 1.46)
16	1.07 (0.82, 1.39)	1.16 (0.88, 1.53)	0.75 (0.51, 1.12)	0.78 (0.53, 1.16)	0.88 (0.65, 1.20)	0.75 (0.58, 0.99)c	1.24 (0.94, 1.64)	1.20 (0.91, 1.57)
17	1.02 (0.77, 1.35)	1.19 (0.90, 1.58)	0.86 (0.59, 1.24)	0.98 (0.70, 1.36)	0.78 (0.582, 1.05)	0.67 (0.48, 0.92)c	1.33 (0.84, 2.12)	1.27 (0.81, 2.00)
18 or Older	0.98 (0.66, 1.46)	1.18 (0.75, 1.86)	0.74 (0.48, 1.16)	0.83 (0.50, 1.37)	0.79 (0.51, 1.22)	0.61 (0.37, 1.03)	2.11 (1.36, 3.27)c	1.91 (1.26, 2.89)c
Race and Ethnicity
Hispanic or Latino (Any Race)	1.38 (1.04, 1.83)c	1.45 (1.08, 1.96)c	1.21 (0.79, 1.86)	1.26 (0.82, 1.94)	1.53 (1.16, 2.02)c	1.72 (1.31, 2.26)c	0.62 (0.50, 0.78)c	0.62 (0.48, 0.80)c
Non-Hispanic Asian	1.00 (0.73, 1.39)	0.98 (0.71, 1.35)	1.12 (0.55, 2.29)	1.01 (0.50, 2.03)	1.28 (0.68, 2.41)	1.26 (0.63, 2.52)	0.40 (0.28, 0.57)c	0.42 (0.28, 0.62)c
Non-Hispanic Black	1.07 (0.78, 1.47)	1.08 (0.77, 1.51)	0.98 (0.58, 1.66)	1.05 (0.60, 1.83)	1.57 (1.10, 2.22)c	1.64 (1.16, 2.33)c	0.45 (0.31, 0.66)c	0.44 (0.30, 0.64)c
Non-Hispanic White (Ref)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Non-Hispanic Other or Multiple Races	1.39 (1.01, 1.90)c	1.33 (0.91, 1.95)	1.42 (0.89, 2.26)	1.32 (0.71, 2.43)	1.71 (1.08, 2.69)c	1.69 (1.05, 2.71)c	0.68 (0.53, 0.89)	0.67 (0.51, 0.89)
Sexual and Gender Identity
LGBTQ+	3.74 (3.01, 4.64)c	2.99 (2.38, 3.77)c	5.13 (4.23, 6.21)c	4.55 (3.76, 5.50)c	2.54 (1.91, 3.37)c	3.02 (2.24, 4.08)c	0.66 (0.54, 0.81)c	0.72 (0.56, 0.91)c
Straight and Cisgender (Ref)	1.00	1.00	1.00	1.00	1.00	1.00	1.00	1.00

Abbreviations: LGBTQ+, Lesbian, Gay, Bisexual, Transgender, Queer or Questioning; YRBS, Youth Risk Behavior Survey.

^aRao-Scott chi-squared test comparing the weighted samples for 2019 and 2021.

^bCounts of 1–4 and associated percentages were suppressed, in line with RIDOH’s Small Numbers Policy.

Of note, in multivariable analyses, female high school students compared to males had higher odds of having experienced persistent feelings of sadness/hopelessness (AOR=2.19, 95% CI=1.88, 2.57) and seriously considered attempting suicide (AOR=1.46, 95% CI=1.28, 1.67) in the past year. LGBTQ+ students had higher odds of having experienced persistent feelings of sadness/hopelessness (AOR=2.99, 95% CI=2.38, 3.77), seriously considered attempting suicide (AOR=4.55, 95% CI=3.76, 5.50), and attempted suicide (AOR=3.02, 95% CI=2.24, 4.08) in the past year, as well as lower odds of having an adult at school with whom they could discuss problems (AOR=0.72, 95% CI=0.56, 0.91), compared to straight and cisgender students. The number of high school students with and without each mental health outcome by year and sociodemographic characteristics is available in Table S2.

In exploratory analyses of potential effect modification, the association between year and each mental health outcome did not differ by race/ethnicity or sexual/gender identity, with two exceptions. First, the association between year and odds of having experienced persistent feelings of sadness/hopelessness differed by race/ethnicity (p=0.02). High school students of color had higher odds of having experienced persistent feelings of sadness/hopelessness in the past year (AOR=1.40, 95% CI=1.19, 1.65) in 2021 compared to 2019. In contrast, non-Hispanic White students in 2019 and 2021 had similar odds of having experienced persistent feelings of sadness/hopelessness in the past year (AOR=1.03, 95% CI=0.82, 1.29). Second, the association between year and odds of having attempted suicide differed by sexual/gender identity (p<0.001). LGBTQ+ high school students in 2019 and 2021 had similar odds of having attempted suicide in the past year (AOR=0.85, 95% CI=0.59, 1.21). However, straight and cisgender students in 2021 compared to 2019 had lower odds of having attempted suicide in the past year (AOR=0.30, 95% CI=0.24, 0.39).

DISCUSSION

In this serial cross-sectional study, middle school students in Rhode Island were more likely to have ever seriously considered and ever attempted suicide during the COVID-19 pandemic (2021) compared to before the pandemic (2019). Among high school students, the likelihood of having felt sad/hopeless in the past year increased among students of color during compared to before the pandemic, while remaining stable among non-Hispanic White students. Additionally, although cisgender and straight high school students were less likely to have attempted suicide in the past year during compared to before the pandemic, this decrease was not observed among LGBTQ+ students (their likelihood of having attempted suicide remained stable). In contrast, the likelihood of high school students having considered suicide in the past year was similar during and before the pandemic across subgroups.

Implications for school health policy, practice, and equity

This study suggests that the COVID-19 pandemic may have worsened multiple aspects of adolescent mental health in Rhode Island, particularly among middle school students. National studies have examined recent mental health trends among adolescents, although studies specific to middle school students are limited. In one study, Garthe et al. found that the prevalence of anxiety and depression symptoms increased among middle school students nationally during (spring 2021) compared to before (fall 2019) the pandemic.¹² Although anxiety and depression symptoms among middle school students were not measured in the present study, the increase in suicidal ideation and suicide attempt is very concerning. Given these trends, it is encouraging that middle school students were more likely to report comfort in seeking non-parental support during compared to before the pandemic, as previous research has noted that school connectedness is protective of mental health.^2,5,8

Findings regarding the impact of the COVID-19 pandemic on the mental health of high school students in Rhode Island were mixed. While we identified an overall increase in the prevalence of feeling sad/hopeless among high school students during the pandemic, in analyses that allowed this trend to vary by race/ethnicity, the increase was only observed among students of color. These findings are consistent with national studies suggesting adolescents of color were more likely to have their mental health impacted by the pandemic and to face pandemic-related stressors such as household unemployment, loss of a loved one, and isolation.¹³ Additionally, although rates of mental health treatment are low for all adolescents, adolescents of color are less likely to receive treatment, and this is often attributed to the lack of a diverse and culturally competent workforce.¹³ Therefore, increasing efforts to recruit and retain a diverse mental health workforce may improve mental health outcomes among adolescents of color. However, in contrast to national studies that have identified an increase in suicidal ideation among all adolescents (i.e., middle and high school students combined) during the pandemic,³ the overall prevalence of seriously considering suicide remained similar and having attempted suicide decreased during the pandemic among Rhode Island high school students. It is possible that the national increase in suicidal ideation among all adolescents was driven primarily by an increase among middle school students, which could be consistent with our findings, or the experience of high school students in Rhode Island may have differed from that of high school students nationally. It is concerning that high school students were less likely to report comfort in seeking non-parental support during compared to before the pandemic, given prior evidence of the importance of school connectedness for mental health.^2,5,8 Nationally, poor mental health was less prevalent among high school students who felt close to people at school,⁵ suggesting that fostering school connectedness may improve mental health. Schoolwide programs focused on social and emotional learning, professional development for staff to improve classroom management, and strategies to promote relationships between students, their families, and school staff may promote school connectedness.⁵

It is important to note that mental health concerns were already on the rise for adolescents nationally and in Rhode Island before the COVID-19 pandemic. Nationally, the percentage of high school students reporting feelings of sadness/hopelessness (26.1% to 36.7%), having seriously considered attempting suicide (13.8% to 18.8%), and having attempted suicide (6.3% to 8.9%) increased from 2009 to 2019.¹⁴ In Rhode Island, the percentage of high school students reporting feelings of sadness/hopelessness increased from 2015 to 2019 (26.4% to 32.3%).¹⁵ Growing use of digital media, increasing academic pressure, limited access to mental health care, alcohol and drug use, and rising income inequality, racism, and gun violence may have contributed to these trends before and during the pandemic.⁶ However, there is some evidence that the pandemic accelerated the increase in mental health concerns among Rhode Island middle and high school students. For example, from 2011 to 2019, the percentage of middle school students who ever seriously considered suicide ranged from 16.9% to 18.7% but, in 2021, increased considerably to 23.2%.¹⁶ Similarly, among high school students, the percentage who felt sad/hopeless had increased from 24.6% to 32.3% between 2011 and 2019, but further increased to 37.8% in 2021.¹⁷

Importantly, sexual/gender minority middle and high school students in Rhode Island experienced a higher prevalence of seriously contemplating suicide, attempting suicide, and feeling sad/hopeless than their heterosexual/cisgender peers. These findings are consistent with previous research suggesting sexual minority adolescents are more likely than heterosexual adolescents to experience poor mental health and suicidal ideation.¹⁴ Trends over time in the present study were generally similar for sexual/gender minority and heterosexual/cisgender adolescents, suggesting these disparities preceded the pandemic and persisted. It is concerning that the likelihood of attempting suicide remained stable from 2019 to 2021 among LGBTQ+ high school students, while their cisgender/heterosexual peers experienced a decline in suicide attempts. Creating safe and supportive school environments (e.g., through student-led clubs that support LGBTQ+ students, school-based positive adolescent development programs, and professional development for staff on policies and practices that support LGBTQ+ students) can reduce suicidal thoughts or attempts among LGTBQ+ adolescents.¹⁸

Female middle and high school students in Rhode Island were also more likely than male students to experience mental health concerns. In middle school, female students were more likely to have ever seriously considered attempting suicide and have ever attempted suicide. In high school, female students were more likely to have felt sad/hopeless and seriously considered suicide in the past year. These findings are consistent with previous work suggesting female high school students are more likely to have felt sad/hopeless, seriously considered attempting suicide, and attempted suicide compared to male counterparts.¹⁴ Although the present study did not examine the impact of the pandemic by sex, some research has found that mental health concerns were more prevalent for females than males during the pandemic,^5,19 which could reflect a disproportionate impact of the pandemic or a pre-pandemic disparity that persisted. This is an important area for future research.

Limitations

This study had important limitations. First, the YRBS question regarding persistent feelings of sadness/hopelessness was changed in the 2021 middle school survey, so we could not examine trends in this measure among middle school students. Additionally, the middle school survey does not measure sexual orientation, limiting consideration of sexual/gender minority middle school students to a measure of whether their gender expression differed from their sex. It is also important to note that, among middle school students, the survey measured whether they had ever seriously considered and attempted suicide rather than recent history. Additionally, YRBS data are self-reported by students and, thus, may be subject to social desirability bias that could result in under-reporting mental health concerns, particularly suicidal ideation and attempt. Finally, due to COVID-19-related delays, the 2019 and 2021 surveys were collected at different times of the year, so differences observed across years may be impacted, in part, by seasonal variation in mental health. However, this study also had strengths. The YRBS is a large and representative survey with a variety of questions related to adolescent health, allowing us to examine multiple mental health measures while controlling for potential confounders.

CONCLUSION

In conclusion, the COVID-19 pandemic may have worsened multiple aspects of adolescent mental health in Rhode Island, particularly among middle school students. High school students of color and those identifying as LGBTQ+ may have experienced disproportionate impacts of the pandemic on mental health. Additionally, LGBTQ+ and female students were more likely to report mental health concerns. Strategies to foster school connectedness, create safe and supportive school environments, and increase the diversity of the mental health workforce may be useful for improving adolescent mental health, generally and within pandemic preparedness efforts. Future research on potential differences in the pandemic impact by student sex and comfort in seeking non-parental support, as well the effectiveness of interventions for improving adolescent mental health, would be useful.