Ecological Factors of Telemental Healthcare Utilization Among Adolescents with Increased Substance Use During the COVID-19 Pandemic: The Moderating Effect of Gender

Abstract

Background

Adolescent substance use is often associated with concurrent mental health problems (e.g., depression, suicide attempts, parental emotional and physical abuse, not feeling close to people at school, and lower virtual connectedness) at multiple ecological levels.

Objective

This study examined whether such risk factors among adolescents were associated with the use of telemental healthcare (TMHC) and whether gender moderated these associations.

Methods

Data were drawn from the Adolescent Behaviors and Experiences Survey, collected by the U.S. Centers for Disease Control and Prevention from January to June 2021. A hierarchical multiple logistic regression analysis was conducted using a national sample of 1,460 students in Grades 9–12 in the United States who reported having used more alcohol and/or drugs during the pandemic than before it started.

Results

The results showed that only 15.3% of students sought TMHC. Students reporting increased substance use during the pandemic were more likely to use TMHC if they experienced more severe mental health problems (e.g., suicide attempts) compared to other ecological factors, such as issues with their family, school, or community. Analysis of the moderating effect showed that the closer male students felt to people at school, the more likely they were to seek TMHC, whereas the opposite was true for female students.

Conclusions

The findings highlighted that feeling close to people at school is an important aspect of understanding the help-seeking behavior of female and male adolescent substance users.

Introduction

The novel coronavirus (SARS-COV-2) outbreak caused by severe acute respiratory syndrome (COVID-19) created unprecedented challenges for adolescents due to abrupt and drastic government measures, such as lockdowns and school closures, upending the protective factors for mental health (e.g., schooling, social support, and community support; Lee 2020). Moreover, the panic surrounding the disease caused a heightened sense of isolation, contributing to profound mental health consequences, including suicide (Gunnell et al., 2020).

Extant data assessing substance use changes among Canadian adolescents from before to during the COVID-19 pandemic showed that the percentage of users of most substances decreased; however, the frequency of both alcohol and cannabis use increased (Dumas et al., 2020). Among a nationally representative sample of students in grades 9–12 (N = 7,705) in the United States from January to June 2021, 33.7% reported past illicit drug (marijuana, synthetic marijuana, cocaine, other illegal drug, or prescription opioid misuse) usage, and 31.4% reported they used more drugs during the COVID-19 pandemic (Brener et al., 2022). Dumas et al. (2020) revealed that most adolescents engaged in solitary use (49.3%) but also used substances with peers via technology (31.6%) or face-to-face (23.6%) during COVID-19.

The COVID-19 pandemic had a significant impact on adolescents’ mental health, likely affecting their substance use behaviors. According to Dumas et al. (2020), high rates of depression and anxiety among adolescents are linked to COVID-19. Among at-risk groups, including those with preexisting mental health conditions or those living in abusive settings, the COVID-19 environment of isolation significantly impaired adolescents’ mental health conditions (Thorisdottir et al., 2021). Consequently, studies documented a high demand for mental health services for adolescents with substance use problems during the pandemic (Czeisler et al., 2020; Racine et al., 2020).

The COVID-19 pandemic caused severe economic issues and social and educational crises (Barron et al., 2021), disrupting existing mental health systems. The National Institute of Mental Health (2022) found that only 40% of adolescents with diagnosed conditions sought help. Lu et al.’s (2021) survey reported that less than 12% of adolescents with co-occurring conditions of major depression and substance use utilized mental health services. Some challenges surrounding adolescents in seeking mental healthcare include mental health literacy (Radez et al., 2021), fear, lack of family support (Al Omari et al., 2022), self and social stigma (Fante-Coleman & Jackson-Best, 2020), and past negative experiences (Gulliver et al., 2010). Although the need for telemental healthcare (TMHC) services (mental healthcare services via telecommunication or videoconferencing; Gulliver et al., 2010; Toscos et al., 2019) has been documented (Tausch et al., 2022) for college students or emerging adults (e.g., Crumb et al., 2021; Toscos et al., 2019), few studies have investigated TMHC frequency among at-risk adolescents who use alcohol and/or drugs and have concurrent mental health issues. What motivates adolescents with increased substance use to seek TMHC is also unclear.

Only a few studies have examined TMHC use among high school adolescents, and none have considered the effect of gender. How gender affected adolescents’ outcomes and their motivation to seek TMHC or other mental health services during the pandemic remains unclear. Equally, the role gender plays in the association between concurrent mental health problems and TMHC utilization among adolescents with increased substance use is uncertain. Although Mackenzie et al. (2006) found that females were more likely than males to seek professional mental health services, no study has hitherto analyzed the role of gender in adolescents seeking TMHC services. We examined how different mental health problems at multiple ecological levels increased adolescents’ willingness to seek TMHC during the COVID-19 pandemic and whether these associations varied by gender.

Theoretical Frameworks

Guided by the ecological systems theory (Bronfenbrenner, 1986) and the health belief model (HBM; Rosenstock et al., 1988), we examined adolescents’ lived environment during the COVID-19 pandemic and their utilization of TMHC to address mental health and substance use. The ecological systems perspective of a person in the environment focuses on the interactions between the individual and their social environment. Bronfenbrenner’s ecological systems theory emphasizes the dependency of living creatures on their surroundings. Healthcare utilization among adolescents, therefore, is the influence of the interaction between the inherent qualities of an adolescent’s perception of the quality of TMHC during COVID-19. This interaction provides synergy with the HBM by predicting adolescents’ beliefs about their mental health conditions and the key factors influencing health-seeking behaviors.

Bronfenbrenner (1979, 1986) divided the person’s environment into five different systems (i.e., the microsystem, mesosystem, exosystem, macrosystem, and chronosystem). Adolescents’ immediate environmental risk factors at the individual level (e.g., depression and suicide attempts), family (parental physical or emotional abuse), school (not feeling close to people at school), and community factors (virtual disconnectedness from family, friends, and other groups) are interactions that may heavily influence their well-being, including mental health, and their beliefs and behaviors (Addison, 1992). However, individuals’ willingness to seek help is a function of the value they assign to the problem (Hutchison, 2019) and their beliefs about the benefits of taking action (Murno et al., 2007; Solanki et al., 2022).

The HBM model, therefore, outlines six constructs that predict health behavior: risk susceptibility, risk severity, benefits to action, barriers to action, self-efficacy, and cues to action (Becker, 1974; Murno et al., 2007; Solanki et al., 2022). The theory assumes that people are likely to avoid a health threat if they believe they have a specific risk of contracting the disease/condition (perceived susceptibility) and consider the situation to pose a serious threat (perceived severity; Rosenstock et al., 1988). Bronfenbrenner’s ecological theory is an effective behavioral model for guiding adolescents’ mental health interventions (Eriksson et al., 2018). It considers individual, family, school, and community in the improvement and sustainability of healthy habits in people (Israel, 1985). However, it does not address health beliefs and associated benefits.

Literature Review

Individual Factors (Depression and Suicidality) and TMH

Toscos et al. (2019) found significant and positive associations between adolescents’ symptoms of depression and suicidality and TMHC help-seeking behaviors. Adolescents reporting elevated symptoms of depression and suicidality had greater use of mental health technology resources (Toscos et al., 2019). Although TMHC, conceptually referred to as teletherapy, telepsychology, or telepsychiatry, has existed for several decades (Hilty et al., 2013; National Institute of Mental Health, 2022), it currently encompasses a broad range of mental health treatment modalities, often offered via online forums, text messaging, or web-based. TMHC has greater accessibility (Fairchild et al., 2020) and applicability (Orsolini et al., 2021) and has been demonstrated to have improved mental health outcomes for adolescents dealing with depression and suicidality than traditional modalities (Calear et al., 2009).

The utility of TMHC was highlighted in a pilot study conducted among children and adolescents in rural emergency departments that found TMHC to be accessible for youth dealing with depression and suicidality and reduced longer emergency department wait times (Fairchild et al., 2020). Likewise, a systematic review of 29 studies focusing on the applicability of TMHC found it was an efficacious alternative to resolving barriers to healthcare delivery among adolescents with depression and suicidality (Orsolini et al., 2021). Another study conducted among students in Grades 9 to 12 (N = 2,789) found slightly over 16% reporting prior experience using TMHC resources, including talking to an online counselor, anonymous chat, using a self-help mobile app, and/or texting a crisis counseling text line (Toscos et al., 2019). Findings reported by Toscos and colleagues (2019) indicated that adolescents with symptoms of depression preferred to have anonymous chats and counseling through crisis text lines, whereas those with suicidality preferred to use all available types of TMHC resources. Thus, the use of TMHC services was contingent on the availability of resources during the COVID-19 pandemic (Racine et al., 2020).

Family Factors (Parental Abuse) and TMHC

Recent research suggests that internet-delivered treatment modalities are viable for adolescents who experience or witness childhood adversity to address child maltreatment, including psychological and physical abuse, neglect, and sexual abuse (Stewart et al., 2021). TMHC removes child maltreatment barriers to treatment access and expands treatment services to diverse populations (Racine et al., 2020). It enhances client treatment adherence (i.e., attendance and engagement), leads to shorter treatment durations, and improves mental health outcomes (Racine et al., 2020). Technology-based trauma treatment (i.e., trauma-focused cognitive behavioral therapy) has been found efficacious in reducing behavioral problems (Comer et al., 2015; Stewart et al., 2017). However, TMHC child maltreatment research remains sparse, with mixed evidence of its implementation or efficacy. Some studies showed a significant but negative relationship between child maltreatment and TMHC (Comer et al., 2015; Stewart et al., 2017), A significant finding, however, is that TMHC treatment modalities for child maltreatment have known limitations, such as ensuring children’s and adolescents’ safety, reliable technologies, and lack of a confidential or private space for processing trauma (Comer et al., 2015; Racine et al., 2020).

School and Community Factors, and TMHC

Research on the association between TMHC and school connectedness (i.e., students feeling valued and having a sense of acceptance), virtual connectedness, and symptoms of mental health is mixed (Jones et al., 2022). Some research has found significant associations between social connectedness and feelings of loneliness, depression, social anxiety, and mental health challenges (Wu et al., 2016). For example, Jones and colleagues (2022) found that adolescents’ virtual connectedness with others (i.e., family, friends, or peer groups via electronic devices) improved their mental health difficulties and lowered their suicidality risk during the COVID-19 pandemic.

However, despite a high prevalence of mental health problems among adolescents, many do not seek professional help (Sheffield et al., 2004). Some prefer to manage their mental independently (Kuhl et al., 1997) or seek help from their friends and families (Sheffield et al., 2004). A majority of adolescents do not want to talk to unfamiliar adults, such as teachers and counselors, about mental health issues; instead, they are more likely to speak to informal sources, such as their friends or family members, to solve their problems (Raviv et al., 2009).

Gender Differences in Ecological Factors and TMHC

The literature demonstrates that adolescent females are more likely to seek mental health services from friends and informal relationships than their male counterparts (Raviv et al., 2009). Literature suggests that the high help-seeking among females is because they may perceive their mental health situation to be more severe (Doherty & Kartalova-O’Doherty, 2020). Further, they may have experienced fewer negative experiences with mental health help-seeking (e.g., stigmatization, fear or adverse outcomes, or difficulty accessing help). Female adolescents are also slightly more likely to have prior TMHC utilization services and knowledge of web-based technologies, anonymous online chats, and self-help resources compared to their male counterparts (Toscos et al., 2019). Yet, some research suggested that male adolescents may have more positive mental health outcomes utilizing TMHC, including reducing depressive symptoms, compared to females (Calear et al., 2009).

Current Study

The circumstances identified in the literature point to various internal and external adolescent risk factors at multiple ecological levels (e.g., individual, relationship, community, and societal) that converge to impact adolescents’ perceptions of mental health and their help-seeking behavior. However, little is known about how these risk factors are associated with high school adolescents’ uptake of TMHC during the COVID-19 pandemic. Further research is warranted to examine individual factors (e.g., depression and suicidality), family factors (e.g., physical and psychological maltreatment), school factors (e.g., feeling connected to others), and community factors (e.g., virtual access to resources and connectedness) that might influence adolescents to seek help—particularly those who perceive that they are susceptible to substance use and mental health issues. Utilizing the ecological systems theory and the HBM, we hypothesized that adolescents would be more likely to seek and experience TMHC if they perceived that they were susceptible to mental health problems or substance misuse. This study examined (a) whether adolescents’ ecological risk factors (depression, suicidal attempts, parental emotional abuse, parental physical abuse, low virtual connectedness, and not feeling close to people at school) and substance use during the COVID-19 pandemic would drive adolescents to use TMHC services and (b) whether these associations would be moderated by gender. Specifically, based on the literature indicating higher rates of help-seeking behaviors among females (Mackenzie et al., 2006; Raviv et al., 2009; Toscos et al., 2019), this study hypothesized that the effects of these ecological risk factors on TMHC use would be greater among female adolescent substance users than their male counterparts during the COVID-19 pandemic.

Methods

Data Source and Sample

The Adolescent Behaviors and Experiences Survey (ABES) conducted among U.S. high school students from January–June 2021 was the data source. The ABES study, an adapted version of the Youth Risk Behavior Surveillance System (YRBS) survey (Underwood et al., 2020), aimed to assess student behaviors and experiences during the COVID-19 pandemic. The survey included 97 questions from the 2021 YRBS questionnaire; six were modified to allow students who were attending only virtually. The ABES survey contains 12 questions on COVID-19–related behaviors that are not included in the YRBS questionnaire.

To obtain a nationally representative sample of public and private school students in Grades 9–12 in all 50 U.S. states and the District of Columbia, a three-stage cluster sampling approach was used. Classes were randomly selected, and data collection was facilitated by teachers who provided the students access to the anonymous online survey. Given the different instructional models used across the nation during the pandemic (i.e., in-person only, virtual-only, and hybrid), the ABES was designed as a self-administered, anonymous 110-item questionnaire that took 30 min on average to complete. A total of 7,998 students submitted surveys, and 7,705 had valid data (i.e., ≥ 20 questions answered). The school response rate was 38%, the student response rate was 48%, and the overall response rate was 18%. The current study used a sample of 1,460 students who reported having used more alcohol and/or drugs during the COVID-19 pandemic than before it started. Specifically, we included students who answered, “strongly agree” and “agree” to the following two questions: (a) Do you agree or disagree that you drank more alcohol during the COVID-19 pandemic than before it started; (b) Do you agree or disagree that you used drugs more during the COVID-19 pandemic than before it started? (Count using marijuana, synthetic marijuana, cocaine, prescription pain medicine without a doctor’s prescription, and other illegal drugs). After deleting cases with missing values, the final sample size for this study was 1,216.

Measures

Dependable Variable

TMHC was measured using the following question: “During the COVID-19 pandemic, did you receive mental healthcare, including treatment or counseling for your use of alcohol or drugs, using a computer, phone, or other device?” We coded these responses in binary form (1 = no and 2 = yes).

Independent Variables

In this study, we included six ecological factors as independent variables: depression (individual factor), suicide attempts (individual factor), parental emotional abuse (family factor), parental physical abuse (family factor), feeling close to people at school (school factor), and virtual connectedness (community factor).

Depression was assessed using the following question: “During the past 12 months, did you ever feel so sad or hopeless almost every day for 2 weeks or more in a row that you stopped doing some usual activities?” The response options were yes and no. These symptoms have frequently been described as depressive (Kim et al., 2018).

Suicide attempts were measured with the question: “During the past 12 months, how many times did you actually attempt suicide?” This measure was assessed on a 5-point Likert scale (1 = 0 times, 2 = 1 time, 3 = 2 or 3 times, 4 = 4 or 5 times, and 5 = 6 or more times), with higher scores indicating more suicide attempts.

Parental emotional and physical abuse were measured using the following questions: (a) “During the COVID-19 pandemic, how often did a parent or another adult in your home swear at you, insult you, or put you down?” (b) “During the COVID-19 pandemic, how often did a parent or other adult in your home hit, beat, kick, or physically hurt you in any way?” The response categories to the questions ranged from 1 (never) to 5 (always). For analysis purposes, the variables were binary coded, reflecting never and rarely as 0 (no) and sometimes, most of the time, and always as 1 (yes).

Virtual connectedness during the pandemic was measured by participants’ responses to the following question: “During the COVID-19 pandemic, how often were you able to spend time with family, friends, or other groups, such as clubs or religious groups, using a computer, phone, or other device?” The response options ranged from 1 (never) to 5 (always).

The feeling of closeness to people at school was assessed with the question, “Do you agree or disagree that you feel close to people at your school?” This measure was assessed on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicated stronger feelings of closeness to people at school.

Moderator

Gender was the moderator in this study, which was a dichotomous variable (0 = male and 1 = female).

Control Variables

We included age and race in the model as control variables. Age comprised five response categories (1 = 14 years or younger, 2 = 15 years, 3 = 16 years, 4 = 17 years, and 5 = 18 years or older). We operationalized race using four response categories (1 = White, 2 = Black/African American, 3 = Hispanic/Latino, and 4 = others).

Data Analysis

Descriptive, bivariate (chi-square tests and independent samples t-tests), and multivariate (hierarchical multiple logistic regression) analyses were conducted using SPSS version 27. The percentages of missing data varied from 0.1% for age to 13.4% for suicide attempts. We omitted cases that had missing data from the analysis. With a large sample size (N = 1,216), list-wise deletion is unlikely to influence the results (Hair et al., 2014). Additionally, before hierarchical multiple logistic regression was conducted, an ordinary least squares (OLS) regression was performed to identify multicollinearity issues among the independent variables. The outcomes of the OLS regression revealed no multicollinearity issues. Specifically, tolerance was > 0.1 and the variance inflation factor was < 10 (Hair et al., 2014).

Finally, we ran a three-step hierarchical multiple logistic regression. Sociodemographic control variables, including age and race, were entered into Model (1) The six ecological factors (depression, suicide attempts, parental emotional abuse, parental physical abuse, feeling close to people at school, virtual connectedness) and the moderator (gender) were then entered into Model (2) The interaction variables between each ecological factor and the moderator were entered into Model (3) The − 2 log-likelihood and Nagelkerke R² values were reported at each step.

Results

Descriptive and Bivariate Results

As reported in Table 1, more than half of the sample (55.1%) self-identified as female. Regarding age, the largest group (28.5%) was 17 years, followed by 16 years (27%), 15 years (21.6%), 18 years or older (15.9%), and 14 years or younger (7.1%). In terms of race, just over half of the sample (52.5%) was White, 12.7% were Black or African American, 4.8% were Hispanic or Latino, and 30.0% of the students were from other racial groups.

Table 1.

Descriptive and Bivariate Results

Variable	Gender n (%)			χ2 or t
	Total 1,216 (100%)	Male 546 (44.9)	Female 670 (55.1)
Age
14 years or younger	86 (7.1)	26 (4.8)	60 (9.0)	15.07**
15 years	263 (21.6)	108 (19.8)	155 (23.1)
16 years	328 (27.0)	144 (26.4)	184 (27.5)
17 years	346 (28.5)	167 (30.6)	179 (26.7)
18 years or older	193 (15.9)	101 (18.5)	92 (13.7)
Race
White	639 (52.5)	293 (53.7)	346 (51.6)	2.25
Black/African American	154 (12.7)	71 (13.0)	83 (12.4)
Hispanic/Latino	58 (4.8)	29 (5.3)	29 (4.3)
Other	365 (30.0)	153 (28.0)	212 (31.6)
Depression
No	399 (32.8)	254 (46.5)	145 (21.6)	84.46***
Yes	817 (67.2)	292 (53.5)	525 (78.4)
Suicide attempts		1.27 ± 0.74	1.41 ± 0.82	-3.13***
0 times	967 (79.5)	467 (85.5)	500 (74.6)	30.34***
1 time	125 (10.3)	30 (5.5)	95 (14.2)
2 or 3 times	93 (7.6)	38 (7.0)	55 (8.2)
4 or 5 times	14 (1.2)	3 (0.5)	11 (1.6)
6 or more times	17 (1.4)	8 (1.5)	9 (1.3)
Parental emotional abuse
No	569 (46.8)	324 (59.3)	245 (36.6)	62.67***
Yes	647 (53.2)	222 (40.7)	425 (63.4)
Parental physical abuse
No	1107 (91.0)	498 (91.2)	609 (90.9)	13.89***
Yes	109 (9.0)	48 (8.8)	61 (9.1)
Virtual connectedness		2.96 ± 1.25	3.08 ± 1.15	-1.79*
Never	153 (12.6)	85 (15.6)	68 (10.1)	13.15*
Rarely	248 (20.4)	116 (21.2)	132 (19.7)
Sometimes	371 (30.5)	144 (26.4)	227 (33.9)
Most of the time	300 (24.7)	138 (25.3)	162 (24.2)
Always	144 (11.8)	63 (11.5)	81 (12.1)
Feeling close to people at school		3.27 ± 1.31	2.95 ± 1.30	4.26***
Strongly disagree	201 (16.5)	77 (14.1)	124 (18.5)	20.33***
Disagree	206 (16.9)	75 (13.7)	131 (19.6)
Not sure	273 (22.5)	122 (22.3)	151 (22.5)
Agree	349 (28.7)	167 (30.6)	182 (27.2)
Strongly agree	187 (15.4)	105 (19.2)	82 (12.2)
Telemental Healthcare
No	1,030 (84.7)	473 (86.6)	557 (83.1)	2.84
Yes	186 (15.3)	73 (13.4)	113 (16.9)

* p < .05, ** p < .01, *** p < .001

Ecological Factors and TMHC by Gender

Table 1 also shows the bivariate results of differences in ecological factors between male and female students. Among the students who used more alcohol and/or drugs during the pandemic, 67.2% reported they had ever been depressed and 20.5% had attempted suicide. Depression and suicide attempts were significantly different between male and female students, χ² = 84.46, p < .001 vs. χ² = 30.34, p < .001, respectively. More female students than male students reported that they ever had depressive symptoms (78.4% vs. 53.5%) or attempted suicide (25.4% vs. 15.5%).

Regarding parental abuse, 53.2% of the respondents reported they had experienced parental emotional abuse and 9.0% had experienced parental physical abuse during the COVID-19 pandemic. Furthermore, parental emotional abuse and physical abuse were significantly associated with gender, χ² = 62.67, p < .001, χ² = 13.89, p < .001, respectively. More females than males experienced parental emotional abuse (63.4% vs. 40.7%) and females and males experienced similar levels of parental physical abuse (9.1% vs. 8.8%).

Among students, 33% reported they were not virtually connected with family, friends, or other groups, and this was significantly associated with gender, χ² = 13.15, p < .05. Specifically, 36.8% of the male students and 29.8% of the female students reported they never or rarely had a virtual connection. Overall, 33.4% reported they did not feel close to people at school, and this was also significantly associated with gender, χ² = 20.33, p < .001. Just over a quarter (27.8%) of the male students and 38.1% of female students reported not feeling close to people at school during the pandemic. In terms of mental healthcare use, 15.3% reported they had ever received TMHC during the pandemic, and 13.4% of the male students and 16.9% of the female students reported they had ever received TMHC during the pandemic. TMHC use did not significantly differ by gender.

Ecological Factors by TMHC

As shown in Table 2, depression was significantly associated with TMHC, χ² = 18.04, p < .001. Among students who experienced depression, 18.4% reported they had ever received TMHC during the pandemic and 81.6% reported they had never received TMHC.

Table 2.

Ecological Factors by Telemental Healthcare (N = 1,216)

Variable	Telemental Health n (%)		χ2 or t
	No 1032 (84.7)	Yes 186 (15.3)
Depression
No	363 (91.0)	36 (9.0)	18.04***
Yes	667 (81.6)	150 (18.4)
Suicide attempts	1.27 ± 0.70	1.74 ± 1.06	-5.79***
0 times	860 (88.9)	107 (11.1)	68.78***
1 time	85 (68.0)	40 (32.0)
2 or 3 times	67 (72.0)	26 (28.0)
4 or 5 times	8 (57.1)	6 (42.9)
6 or more times	10 (58.8)	7 (41.2)
Parental emotional abuse
No	507 (89.1)	62 (10.9)	15.98***
Yes	523 (80.8)	124 (19.2)
Parental physical abuse
No	951 (85.9)	156 (14.1)	13.82***
Yes	79 (72.5)	30 (27.5)
Virtual connectedness	3.04 ± 1.19	2.98 ± 1.21	0.65
Never	130 (85.0)	23 (15.0)	2.96
Rarely	204 (82.3)	44 (17.7)
Sometimes	314 (84.6)	57 (15.4)
Most of the time	262 (87.3)	38 (12.7)
Always	120 (83.3)	24 (16.7)
Feeling close to people at school	3.15 ± 1.28	2.80 ± 1.43	3.08***
Strongly disagree	152 (75.6)	49 (24.4)	20.72***
Disagree	169 (82.0)	37 (18.0)
Not sure	243 (89.0)	30 (11.0)
Agree	307 (88.0)	42 (12.0)
Strongly agree	159 (85.0)	28 (15.0)

*** p < .001

Likewise, suicide attempts were significantly associated with TMHC. According to the independent samples t-test, students who had never received TMHC during the COVID-19 pandemic showed lower suicide attempts (M = 1.27) than students who had ever received TMHC during the pandemic (M = 1.74), t = -5.79, p < .001. The chi-square test also showed a significant association between suicide attempts and TMHC, χ² = 68.79, p < .001. Among students who attempted suicide only one time, 32.1% reported they ever received telemental health. Among those who attempted suicide two or three times, 28% reported they had ever received TMHC during the pandemic. Among those who attempted suicide four or five times, 42.9% reported they had ever used telemental health. Among students who attempted suicide six or more times, 42.1% reported they had ever received TMHC.

Parental emotional abuse and physical abuse were also significantly associated with TMHC, χ² = 15.98, p < .001, χ² = 13.82, p < .001, respectively. Among students who had experienced parental emotional abuse, 19.2% reported they had utilized TMHC. Among students who had experienced parental physical abuse, 27.5% reported they had received TMHC. However, virtual connectedness and TMHC were not significantly associated.

Feeling close to people at school was inversely associated with TMHC. According to the independent samples t-test, students who had never received TMHC during the COVID-19 pandemic showed higher levels of feeling close to people at school (M = 3.15) than students who had ever received TMHC during the pandemic (M = 2.80), t = 3.08, p < .001. The chi-square test also showed a significant association between feeling close to people at school and TMHC, χ² = 20.72, p < .001. Among students who had never felt close to people at school, 24.4% reported they had received TMHC and among those who rarely felt close to people at school, 18.1% reported they had used TMHC during the pandemic.

Correlations

Table 3 provides the correlations between ecological factors and TMHC. Depression, r = .12, p < .001, suicide attempts, r = .21, p < .001, parental emotional abuse, r = .12, p < .001, and parental physical abuse, r = .11, p < .001, were all positively correlated with TMHC. However, feeling close to people at school was negatively correlated with TMHC, r = − .10, p < .001.

Table 3.

Correlations

	1	2	3	4	5	6	7	8	9	10	11	12
1. Age	1
2. White	0.02	1
3. Black/African American	0.03	− 0.40***	1
4. Hispanic/Latino	− 0.03	− 0.24***	− 0.09**	1
5. Female	− 0.11***	− 0.02	− 0.01	− 0.02	1
6. Depression	− 0.04	− 0.01	− 0.09***	0.01	0.26***	1
7. Suicide attempts	− 0.10***	− 0.05	0.03	− 0.04	0.09**	0.28***	1
8, Parental emotional abuse	− 0.10***	− 0.07*	− 0.03	0.00	0.23***	0.35***	0.23***	1
9. Parental physical abuse	− 0.11***	− 0.07*	0.02	− 0.00	0.01	0.01***	0.34***	0.27***	1
10. Virtual connectedness	− 0.03	0.04	− 0.01	− 0.03	0.05	− 0.02	− 0.04	0.00	− 0.05	1
11. Feeling close to people at school	− 0.06*	0.06*	− 0.05	− 0.00	− 0.12***	− 0.22***	− 0.10***	− 0.14***	− 0.05	0.11***	1
12. Telemental healthcare	− 0.02	0.04	− 0.05	− 0.02	0.05	0.12***	. 21***	0.12***	0.11***	− 0.02	− 0.10***	1

* p < .05, ** p < .01, *** p < .001

Hierarchical Multiple Logistic Regression Results

Table 4 presents the results of the multivariate hierarchical logistic regression analysis.

Table 4.

Hierarchical Multiple Logistic Regression of Telemental Healthcare

Variable	Model 1		Model 2		Model 3
	OR	95% CI	OR	95% CI	OR	95% CI
Age	0.95	(0.83, 1.09)	1.02	(0.89, 1.17)	1.01	(0.88, 1.16)
White	1.08	(0.76, 1.53)	1.24	(0.86, 1.78)	1.24	(0.85, 1.79)
Black/African America	0.63	(0.35, 1.14)	0.65	(0.35, 1.20)	0.64	(0.35, 1.19)
Hispanic/Latino	0.74	(0.32, 1,71)	0.86	(0.36, 2.04)	0.81	(0.34, 1.96)
Female			1.05	(0.74, 1.48)	1.61	(0.38, 6.91)
Depression (D)			1.35	(0.87, 2.10)	1.00	(0.54, 1.84)
Suicide attempts (SA)			1.59	(1.32, 1.92) ***	1.96	(1.45, 2.66) ***
Parental emotional abuse (PEA)			1.33	(0.92, 1.94)	1.49	(0.82, 2.68)
Parental physical abuse (PPA)			1.25	(0.74, 2.12)	1.41	(0.60, 3.30)
Virtual connectedness (VC)			0.99	(0.86, 1.14)	0.87	(0.70, 1.07)
Feeling close to people at school (FCPS)			0.87	(0.77, 0.99) *	1.03	(0.85, 1.26)
D × Female					1.77	(0.70, 4.47)
SA× Female					0.73	(0.49, 1.07)
PEA × Female					0.84	(0.39, 1.81)
PPA × Female					0.82	(0.27, 2.44)
VC × Female					1.25	(0.94, 1.65)
FCPS × Female					0.74	(0.57, 0.96) *
-2 Log likelihood	1035.45		974.19		962.20
Nagelkerke R2	0.01		0.09		0.11

* p < .05, *** p < .001. Note. Reference categories – male and race-other

First, we entered the demographic control variables into Model 1 but found none to be significant. Next, the six ecological variables (depression, suicide attempts, parental emotional abuse, parental physical abuse, feeling close to people at school, virtual connectedness) and the moderator (gender) were entered into Model 2 to examine the association between ecological factors and TMHC. All demographic control variables remained nonsignificant in Model 2, except for suicide attempts and feeling close to people at school, which were significantly associated with TMHC. For every additional category of suicide attempt, the odds of TMHC usage were increased by 59% (OR = 1.59, 95% CI [1.32, 1.92]). Likewise, for every additional category of feeling close to people at school, the odds of TMHC usage were decreased by 13% (OR = 0.87, 95% CI [0.77, 0.99]).

Lastly, the interaction variables were entered into Model 3, as shown in Table 4, to examine the moderating effect of gender on the associations between ecological factors and TMHC. In Model 3, only suicide attempts (OR = 1.96, 95% CI [1.45, 2.66]) remained significant. In Model 3, the moderating effect of gender on the association between feeling close to people at school and TMHC was significant (OR = 0.74, 95% CI [0.57, 0.96]). As illustrated in Fig. 1, the association between feeling close to people at school and TMHC was positively associated with being male, but it was negatively associated with being a female student. When male students felt close to people at school, they were more likely to access TMHC. On the other hand, when female students felt close to people at school, they were less likely to access TMHC.

Fig. 1.

Moderating Effect of Gender on the Association Between Feeling Close to People at School and Telemental Healthcare

Discussion

The present study focused on adolescents with increased substance use during the COVID-19 pandemic and the ecological factors (i.e., depression, suicide attempts, parental emotional and physical abuse, virtual connectedness, and feeling close to people at school) related to their use of TMHC and examined gender as a moderating factor.

The descriptive results showed that only 15.3% of adolescents sought TMHC (84.7% did not) and that 13.4% of males and 16.9% of females sought TMHC. These results support those of previous studies and signify a potentially low level of perceived susceptibility to mental health risks among the sample and a low level of perceived benefit of TMHC use (Toscos et al., 2019). The TMHC use rates were low among those who had further identified ecological risk factors; 67.2% reported experiencing symptoms of depression, but only 18.4% reported receiving TMHC. More than one in five adolescents in our sample reported having attempted suicide at least once but less than half sought TMHC. However, these adolescents were more likely to seek TMHC than those who did not experience these factors, supporting the results of previous studies and the proposed theoretical framework, which posits that intrapersonal and interpersonal factors in the lives of adolescents influence their likelihood of seeking healthcare (Pater et al., 2020; Toscos et al., 2019). Adolescents who experienced parental emotional and physical abuse were also more likely to receive TMHC, with 19.2% of those who experienced emotional abuse and 27.5% of those experiencing physical abuse receiving TMHC (Racine et al., 2020).

Overall, the results of our study partially support our first hypothesis; suicide attempts were significantly associated with TMHC use, with the likelihood of seeking care increasing with the number of suicide attempts made. Adolescents with increased substance use during the pandemic showed more health-seeking behaviors when they had more severe mental health problems (e.g., suicide attempts) than other family, school, or community risk factors. This reflects perceived severity in the HBM, where the experience of making suicide attempts strengthens one’s belief about the severity of the existing threat (Rosenstock et al., 1988), which might have driven one to attempt suicide. Thus, it must be ensured that crisis services, such as TMHC, are available whenever adolescents need them. Suicide screening and assessment should also be implemented for adolescents with substance use problems.

Feeling close to people at school was also associated with TMHC use. The closer adolescents felt to people at their school, the less likely they were to seek TMHC. Adolescents with increased substance use during the COVID-19 pandemic were less likely to use TMHC when they felt close to their peers at school. An increase in school connectedness results in lower susceptibility to mental health issues. This is an interesting finding because previous studies have found an increase in school connectedness among adolescents results in lower susceptibility to mental health issues, suggesting, based on HBM, that such adolescents are less likely to seek TMHC (Jones et al., 2022). Generally, adolescents are less likely to talk to authoritative figures, such as counselors and teachers, and more likely to speak to informal sources, such as their friends or family members (Raviv et al., 2009). Adolescents tend to be uncomfortable speaking to unfamiliar adults (Helms, 2003). When adolescents have close peers to talk to about their substance or mental health problems, they may be less likely to seek TMHC.

Our study contributes to the literature by including the moderating effect of gender. Our results partially support our second hypothesis concerning the moderating effect of gender on associated ecological factors and TMHC use. Specifically, we found the association between feeling close to people at school and TMHC varied depending on gender: the closer male students felt to people at school, the more likely they were to seek TMHC, whereas the closer female students felt to people at school, the less likely they were to seek TMHC. Positive support from significant relationships has been found to increase formal help-seeking for mental health in males (Harding & Fox, 2015). This could be because men have attributed close relationships to cultivating confidence and trust in seeking formal mental healthcare (Burke et al., 2022; Vogel et al., 2007). According to Raviv et al. (2009), female adolescents are less likely to seek counseling because they are more likely than male adolescents to open up to peers about their feelings and mental health problems, such as depression or anxiety. However, previous studies have shown that female adolescents are slightly more likely than male adolescents to have previous experience with TMHC resources and prefer to use various types of web-based technologies, with a higher preference for anonymous online chats and self-help resources, when seeking help for their mental health symptoms (Toscos et al., 2019). Our study findings highlight that feeling close to people at school is an important factor in understanding the help-seeking behaviors of female and male adolescent substance users.

Limitations

Despite the important findings of this present study, it had several limitations. First, while the study examined a nationally representative sample of high school adolescents, the data were cross-sectional, which restricted our ability to describe causal interpretations of the associations between the independent variables (depression, suicide attempts, parental emotional and physical abuse, virtual connectedness, and feeling close to people at school) and the dependent variable (TMHC use).

A further limitation concerns the utilization of self-reported data on substance use. Some participants may not have accurately reported their current substance use due to the social stigma associated with it (Kim et al., 2018). Additionally, rather than examining all adolescents who reported current substance use, the present study examined only adolescents who reported increased substance use during the COVID-19 pandemic. Future research should consider including all current substance users as a sample.

Next, depression was operationalized as a single variable from the secondary dataset, which may have weakened the contentment validity of the measure or omitted some depressive symptoms. Some previous studies have also equated “sadness and hopelessness” with depression (Messias et al., 2014; Reed et al., 2015). Furthermore, more nuanced information regarding the type of TMHC used and the frequency of TMHC use would have provided a deeper understanding of perceived barriers, a construct of the HBM that was not explored in depth in this present study (Solanki et al., 2022).

Lastly, this study focused on more micro-level ecological factors that have immediate impacts on TMHC use. Future research should also consider mezzo- and macro-level ecological factors associated with TMHC use by adolescents with substance use issues. Other important factors that may influence TMHC use, such as family economic status, family structure, health insurance, and community characteristics (rural or urban), could be considered for future research.

Implications and Conclusions

Given that 15.3% of students who reported increasing their substance use during the COVID-19 pandemic also sought TMHC during the same period, strategies to increase the utilization of mental healthcare must be considered. Schools are the largest source of mental healthcare for adolescents (Merikangas et al., 2011), this gives school personnel, including teachers, administrators, and mental health practitioners, an opportunity to reduce stigma and increase access to mental health services within the school environment (Wolpert & Cortina, 2018). Importantly, norms can be established regarding substance use and mental healthcare through conversation, education, and service dissemination (Probst et al., 2021). Incorporating mental health discussions into social-emotional curricula and health services can promote service utilization by normalizing mental healthcare. Additionally, facilitating dialogue and regularly sharing knowledge of educational resources for mental health can reduce stigma and increase help-seeking behaviors, cultivating a climate of acceptance (Boydell et al., 2013).

With a recent increase in adolescent usage of mobile apps for mental health and substance use recovery, implementing practices that expand TMHC may also increase mental healthcare usage for adolescents. Adams et al. (2021) conducted a mixed-methods study that found that adolescents who had both substance use and mental health needs preferred using an mHealth app over paper documentation as part of treatment. Adolescents were drawn to these platforms because of their convenience, accessibility, and capacity to deliver immediate feedback (Adams et al., 2021). Additionally, many resources to help clinical practice were developed amid the COVID-19 pandemic that may help identify resources on TMHC provided by professional organizations (Briere et al., 2020).

Prominent policies (e.g., the Every Student Succeeds Act and No Child Left Behind) did increase funding for mental healthcare to be provided in schools for adolescents (ESSA, 2015; Flaherty & Orsher, 2002). States have also implemented laws that increase funding to hire more school mental health professionals and reimburse TMHC services. However, statistics on the youth mental health crisis during the COVID-19 pandemic have made it clear that many schools still experience barriers in meeting the mental health needs of their students, with the most significant barriers being inadequate funding and lack of access to licensed mental health professionals (Padgett et al., 2020). Most recently, the American Rescue Plan and the Bipartisan Safer Communities Act provided billions of dollars for states to implement and expand school-based mental health and substance use services (U.S. Department of Education, 2022). Future policies increasing access to TMHC could include strategies for reducing barriers to care, such as expanding health insurance policies to reach more adolescents and increasing internet access in resource-scarce areas (Schwarz & Aratani, 2011). Future studies should seek to identify barriers that need to be remediated to increase adolescent usage of TMHC. Since mental health issues and substance abuse often co-occur, this is a critical issue that must be urgently addressed.

Due to differences between male and female adolescents in help-seeking behaviors and social issues, it is important to consider differentiated treatment strategies and screening tools (Lu et al., 2021). For male adolescents, mentoring, group interactions, and shared activities in informal spaces with a focus on expressing emotions and experiences of mental health may promote seeking mental healthcare (Burke et al., 2022; Calear et al., 2009). To mitigate feelings associated with the stigma of mental health or substance use, self-administered screening tools may be more beneficial for female adolescents (Substance Abuse and Mental Health Services Administration, 2009). In sum, our study provides novel insight into potential gender differences concerning how feeling close to people at school relates to TMHC use among adolescents. Our findings highlight the need for future research that explores these gender differences in greater detail and helps to inform an evidence base that will impact treatment efforts for adolescents with substance and mental health problems.

Declarations

Conflict of Interest

This study was not supported by any funding and the authors have no conflicts of interest to declare.

Access to Data

ABES is readily available and furnished through the U.S. Centers for Disease Control and Prevention (CDC). All authors equally take responsibility for the integrity and accuracy of the data analysis.