Sexual Violence and Substance Use Among Portuguese College Students: Mental Health Implications

Abstract

Background: Research shows that experiences of sexual violence and the use of substances hurt mental health, predisposing individuals to the development of psychological symptoms of somatization, depression, and anxiety. Objective: The main objective of this study is to assess substance use, sexual violence, and psychological symptoms (BSI-18) in a community sample of Portuguese college students. Participants: 459 college students aged 18 to 50 years participated in this study (mean = 21.55; SD = 3.15). Methods: A survey was created online and disseminated via mailing lists and social networks, and the measurement instruments used were an informed consent form, a sociodemographic questionnaire, the Brief Symptom Inventory-18 (BSI-18), a Sexual Violence Questionnaire, and a Substance Use Questionnaire. Results: We found that men consume more substances, but women report more sexual victimization and psychological symptoms. Non-heterosexuals report significantly more sexual violence experiences and higher substance consumption and psychological symptoms than heterosexuals. Sexual victimization showed strong correlations with both substance use and psychological symptoms, while only some substances correlated with psychological symptoms. Age, gender, sexual orientation, sexual violence, and the substance “Hashish/Marijuana” were shown to be significant and strong predictors of psychological symptoms. Conclusions: As shown in previous studies, sexual and gender minorities are more at risk of experiencing sexual victimization and are more prone to consuming substances and developing psychological symptoms, except for men, who tend to consume more substances. Sexual violence and psychological symptoms are highly correlated, while the relation between substance use and psychological symptoms is not entirely clear. Our findings highlight the need for stronger educational and clinical interventions and increased mental health resources, with training on topics involving sexual violence, substance use, and addressing the specific challenges faced by minority groups, to create safer, inclusive, and more supportive Portuguese universities.

1. Introduction

Psychological symptoms (PSs) are those related to psychological dysfunction, and their presence may be a sign of emotional distress or, if very evident, an indication of psychological disorders. These may manifest mainly through symptoms of anxiety, depression, and somatization that involve the manifestation of physical symptoms such as fainting, dizziness, nausea, and hyperventilation due to psychological factors (Canavarro, 2007; Nazaré et al., 2017) and may be associated with several factors such as personality traits, negative behaviors, or stressful life events.

These symptoms seem to be more apparent when it comes specifically to college and university students, possibly since these students are exposed to specific factors such as adaptation to living independently and away from parents (Ishii et al., 2018); concerns related to academic performance like competition and comparison to peers, fear of disappointing parents/caretakers due to low grades or choice of career paths, heavy workload and other types of concerns (Erschens et al., 2019; Ishii et al., 2018; Ratanasiripong et al., 2018); disrupted sleep or eating patterns as a consequence (Trigueros et al., 2020) and a general lack of self-care (Jenkins et al., 2019; Rosenthal et al., 2018); a lack of balance between academics and social life (Ishii et al., 2018); a low-income background (Rosenthal et al., 2018; Usher & Curran, 2019); underlying psychopathological conditions and the associated societal stigma (Ishii et al., 2018; Park et al., 2020); experienced sexual violence or victimization (SV) (McDougall et al., 2019; Rosenthal et al., 2018); substance use (Rosenthal et al., 2018); and more recently, the COVID-19 pandemic (Cao et al., 2020; Kecojevic et al., 2020; Maia & Dias, 2020; Wang & Zhao, 2020), among many others.

According to the Minority Stress Theory (Meyer, 2003), minority groups, such as sexual and gender minorities (SGMs) (Sexual and Gender Minority Research Office [SGMRO], 2019), are more prone to suffer from PSs due to prejudice and associated stigmas. Thus, SGMs may also be more prone to experience Sexual Violence (SV) or to consume more or heavier substances (Hatzenbuehler, 2009; Neves et al., 2023; Katz et al., 2025; Dyar, 2025) to regulate their emotions, conform to social norms around consumption, or meet expectations of increased sociability and reduced tension while consuming (Hatzenbuehler, 2009).

Substance use (SU) has become an increasingly prevalent health issue among college students globally (Durowade et al., 2021). Alcohol, tobacco, and cannabis are the substances most widely used and consequently the most represented in the literature. The reasons behind statements of consuming licit or illicit substances vary, such as coping with mental stress or disorders, addiction, disinhibition, socialization, bonding and integration in a group (Colby et al., 2009; McHugh et al., 2018), sensation seeking, personality structure (e.g., addictive personality), drug expectancy, and social and environmental factors (Maisto et al., 2019). Therefore, it is important to consider the biopsychosocial model, which proposes that biological, psychological, and sociocultural factors jointly contribute to substance use (Skewes & Gonzalez, 2013).

It is still unclear how SU and PSs are related. Some authors say that depressive symptoms are associated with tobacco, cannabis, cocaine, amphetamine, sedative, and hallucinogen use, but not alcohol nor designer drug use; anxiety symptoms are not associated with the use of any substance (Walters et al., 2018). Others say depression disorders are also connected to alcohol use and anxiety disorders to cannabis consumption (Blanco et al., 2018).

During the beginning of the COVID-19 pandemic, there was a worrisome increase in PSs, especially in sexual and gender minorities (Chaiton et al., 2021). Specifically, substances such as tobacco, alcohol, and cannabis were more frequently used when assessing poorer mental health (Chaiton et al., 2021; Salerno et al., 2021). This escalation in consumption may have happened due to coping mechanisms used to deal with the stress caused by the pandemic, noting that there may be symptoms of depression in individuals who used tobacco, alcohol, or both as coping mechanisms (Martínez-Cao et al., 2021). However, not all studies support either position (e.g., Charles et al., 2021).

In Portugal, during 2016 and 2017, alcohol (86.4%) was the most commonly consumed substance, followed by tobacco (48.8%), medications (12.1%) and illicit psychoactive substances (11.7%), mainly cannabis (11%), cocaine (1.2%), and ecstasy (0.7%) (Balsa et al., 2018). Among these, consumption was higher in men, except for medications. Young adults (15–34 years old) had a higher prevalence of experimental use of tobacco (52.7%) and illicit psychoactive substances (16%) such as cannabis (15.1%), ecstasy (0.9%), LSD (0.5%), and new psychoactive substances (0.5%) compared to the general population (15–64 year olds) (Balsa et al., 2018). Also in Portugal, a different study focusing specifically on young adults who were university students and their experimentation with substances found that tobacco (57.7%), alcohol (44.2%), cannabis (35.3%) were the most experimented substances, with men showing higher rates of experimenting with tobacco (Reis et al., 2017). There were no significant differences between men and women when it came to alcohol and cannabis; however, it was also found that women tended to consume drugs for the first time more frequently, but men had a more frequent regular drug consumption (Reis et al., 2017).

Most studies indicate that men tend to consume substances more often, in greater quantities, or both compared to women (Blanco et al., 2018; SICAD, 2022; Phillips, 2025). However, women who use substances often manifest higher levels of PSs (Di Paola et al., 2022; Gao et al., 2020). Similar to what occurs in sexual victimization, transgender and non-binary individuals are the ones in whom SU recurs more frequently (Hughto et al., 2021; Newcomb et al., 2020). Concerning sexual orientation, studies found that bisexuals, especially cisgender women (Schuler & Collins, 2020), are at a higher risk of SU comparing to other sexual minorities, followed by cisgender homosexual individuals, and cisgender heterosexuals are the ones less at risk (Coulter et al., 2017; Krueger et al., 2020; Schuler & Collins, 2020).

According to the Portuguese Association for Victim Support (Associação Portuguesa de Apoio à Vítima [APAV], n.d.), sexual violence can be defined as any unwanted or non-consensual sexual action forced upon individuals, either attempted or consummated, such as intimate touches or kisses, comments or jokes, threats or aggression, and rape or attempted rape.

College students are more vulnerable to suffering SV (Bhochhibhoya et al., 2021; Dilip & Bates, 2021). The Portuguese academic context is characterized by diverse academic celebrations throughout the year (e.g., “praxe”—freshman rituals and reception; “queima das fitas”—celebration of the end of the academic year; and “semana académica”—a week of musical concerts). This environment can lead to gender inequalities, the normalization of sexual violence, substance use, and the reproduction of myths about sexual violence through preventive educational actions based on sexism (Pires et al., 2018).

In addition, college students are exposed to multiple risk factors for both perpetration and victimization at the individual level (e.g., race and ethnicity, gender, sexual orientation, among others) and the community, contextual, and institutional levels (Bonar et al., 2022). Cisgender women are more at risk of becoming SV victims when compared to cisgender men (Coulter et al., 2017; Gómez, 2022; Ortensi & Farina, 2020; Bhochhibhoya et al., 2021). Specifically, LGBTQIA+ individuals are more likely to suffer SV when compared to heterosexual cisgender individuals, with bisexual and transgender individuals being the most likely to suffer from SV and harassment (Coulter et al., 2017; Martin-Storey et al., 2018; Ortensi & Farina, 2020; Snyder et al., 2018).

Given that these are intense and traumatic situations, there is a clear correlation between having experienced SV and having poorer mental health (Binion & Gray, 2020; Dilip & Bates, 2021; Gómez, 2022). Symptoms or diagnoses of anxiety and depression, among other mental health disorders, are significantly associated with SV experiences, (Depraetere et al., 2023; Tarzia et al., 2020).

The main objective of this study is to assess SU, SV, and PSs levels in a community sample of Portuguese college students. This was split into three specific objectives: (1) compare the differences between levels of SU, SV, PSs and sociodemographic variables (gender and sexual orientation); (2) assess the degree of association between SU and SV, SU and PSs, and SV and PSs; and (3) determine the predictive power of sociodemographic variables, SU, and SV on PSs. For this purpose, sociodemographic variables, SV, and SU were considered independent variables, while PSs were the dependent variables.

Studies on the relationships between substance use, sexual violence, and psychological symptoms in Portugal are scarce, with most studies investigating one of these variables in isolation. Considering that academic environments often present high levels of substance use, sexual violence (Burke et al., 2025), and, consequently, psychological symptoms, it is relevant to analyze these factors in an integrated procedure, especially at a time when Portugal is experiencing demographic, political, and economic changes that generate highly adverse social impacts, such as unemployment, inequality, and poverty (Caleiras & Carmo, 2024).

2. Materials and Methods

2.1. Participants

A total of 459 college students participated in this study, whose ages ranged from 18 to 50 years, the mean age was 21.55 years, the standard deviation was 3.15 years, the median was 21 years, and the mode was 22 years. Most of the participants were young adults, since about 90% were under the age of 23. Concerning gender, most participants self-identified as women (72.1%), 27% as men, and only 4 participants identified as non-binary (0.9%). As for sexual orientation, the majority identified as heterosexual (79.8%), 11.4% as bisexual, 17 participants as homosexual (3.7%), and 16 pansexual (3.5%), only 4 participants answered asexual (0.9%), and 3 chose the alternative “other” (0.7%). This and other sociodemographic characteristics can be seen in

Table 1. Sociodemographic characteristics (M_age = 21.55, SD = 3.15).

		n	%
Gender	Woman	331	72.1
	Man	124	27
	Non-binary	4	0.9
Citizenship	Portuguese	394	85.8
	Brazilian	55	12
	Other	10	2.2
Marital Status	Single	324	70.6
	Married	4	0.9
	Separated/Divorced	1	0.2
	Non-Marital Partnership	2	0.4
	Significant Relationship/Dating	128	27.9
Academic Qualifications	12 years	203	44.2
	Degree/Bachelor Degree	230	50.1
	Post-Graduation/Master’s Degree	23	5
	PhD/Postdoctoral	1	0.2
	Other	2	0.4
Place of Residence	Small Rural Area	116	25.3
	Large Rural Area	44	9.6
	Small Urban Area	205	44.7
	Large Urban Area	94	20.5
Socio-Economic Status	Low	17	3.7
	Low–Medium	129	28
	Medium	259	56.5
	Medium–High	53	11.6
	High	1	0.2
Sexual Orientation	Heterosexual	366	79.8
	Bisexual	53	11.4
	Homosexual (Gay or Lesbian)	17	3.7
	Asexual	4	0.9
	Pansexual	16	3.5
	Other	3	0.7
Professional Situation	Student	442	96.3
	Student Worker	17	3.7

.

2.2. Measurement Instruments

This study will use a quantitative research design, employing inferential, correlational, and predictive analyses to examine the relationships between variables.

The informed consent form was constructed to enlighten the participants on the objectives of this study, the entity responsible, the time required to complete the questionnaire, and contact information, among other aspects. In addition, concepts like anonymity, confidentiality, personal data protection, and the principle of volunteering were clarified and guaranteed.

The sociodemographic questionnaire was constructed with 9 items to assess age, gender, citizenship, marital status, academic qualifications, place of residence, socio-economic status, sexual orientation, and professional situation.

The Brief Symptom Inventory-18 (BSI-18) was developed by Derogatis (2000) to assess psychological distress in clinical and community contexts; the instrument evaluates three dimensions: somatization, depression, and anxiety. It consists of 18 items, 6 for each of 3 subscales. The somatization subscale includes 01. Faintness or dizziness; 04. Heart or chest pains; 07. Nausea or stomach discomfort; 10. Difficulty breathing; 13. Sensation of numbness or tingling in parts of your body; and 16. Feeling weakness in parts of your body. The depression subscale consists of the items 02. Lack of interest in things; 05. Feeling lonely, 08. Feeling sad, 11. Feeling worthless, 14. Feeling hopeless about the future, and 17. Life-ending thoughts. And the items for the anxiety subscale are 03. Nervousness or inner agitation; 06. Feeling tense or nervous; 09. Getting scared suddenly for no reason; 12. Moments of terror or panic; 15. Feeling so agitated you cannot sit still; and 18. Feeling afraid. The mean scores of these dimensions can be used to derive a global indicator of psychological symptoms. It has a Likert-type response format from 0 (Never) to 4 (Always), and psychological symptoms comparable to the normative population can be calculated, thus allowing the assessment of symptom levels (for example, high or low levels). Cronbach’s alpha was 0.92, indicating excellent reliability (Nazaré et al., 2017). Additionally, to understand whether the responses given were exacerbated by the COVID-19 pandemic, a question with an answer format from 0 to 10 was added (“How were the answers you gave to the previous questions increased by the COVID-19 pandemic?”).

For the Sexual Violence Questionnaire, the questions were designed to address the objectives of this study and to assess the occurrence of sexual violence, based on a prior pilot study. It consists of 12 items with an answer format from 0 to 10, focusing on both the victim and the perpetrator. Questions directed to victims were as follows: “Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?”, “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?”, “Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?”, “Has anyone ever kissed or touched you sexually without your consent?”, “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?”, and “Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?”. For perpetrators, the questions were as follows: “Have you ever forced someone to have sexual intercourse against their will, knowing that, under the influence of alcohol or drugs, that person would not have the strength to resist?”, “Have you ever forced someone to have sexual intercourse sex against their will, knowing them well (friend, boyfriend/girlfriend, acquaintance, etc.)?”, “Have you ever initiated sexual contact with someone, involving penetration (oral, vaginal, or anal) without their consent?”, “Have you ever kissed or touched someone sexually without their consent?”, “Have you ever sent an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures or videos to someone, knowing it would make them feel uncomfortable or offended?” and “Have you ever made sexual comments about someone or told jokes or stories of a sexual nature about them that you know would make them uncomfortable or offended?”.

The Substance Use Questionnaire was also created to describe the consumption frequency of the most common substances. It consists of 12 items with 5 response options from “Never” to “6/7 times” per week. The 12 items are as follows: “Cigarettes or other tobacco”, “Electronic cigarette or vaping”, “Alcoholic beverages (beer, wine, etc.)”, “Alcoholic beverages (shots, spirits, etc.)”, “Hashish/Marijuana”, “Unprescribed medication”, “Cocaine or crack”, “Heroin”, “Methamphetamines”, “Magic mushrooms (hallucinogens)”, “Ecstasy/Molly/MDMA” and “Other drugs”. At the end, a dichotomous question was added (with response options “Yes” or “No”) to assess whether participants self-assessed excessive consumption (“For each of the substances you marked, do you consider it was excessive consumption?”).

2.3. Procedures

The sample was collected through convenience sampling with a survey built on an online website and disseminated via mailing lists and social networks. This survey was online from 27 September to 30 October 2022. The survey was disseminated through main lists, social media, and student groups. The invitation message included information about ethical aspects, such as informed consent, inclusion criteria, clarification of this study’s objectives, assurance that participants could voluntarily disclose personal data for the research confidentially and anonymously by guaranteeing personal data protection (encrypted IP), and the volunteering principle, meaning only people willing to answer became participants, and that this study was approved by the Ethics Committee of University of Beira Interior. Subsequently, data stored in EXCEL 365 files were used to build a database in the IBM SPSS Statistics program, version 28.0, to perform the planned statistical evaluations.

2.4. Data Analysis

Firstly, participants were eliminated from the SPSS (v.28) database after considering the exclusion criteria. To proceed with the data analysis, clustering some categories into two variables was necessary: in sexual orientation, the categories “bisexual”, “homosexual”, “asexual”, “pansexual” and “other” were clustered into a single category named “non-heterosexual”. Given the small total sample size of 96 individuals, it was not feasible to analyze the complexity of these identities separately, and they were combined into a single cluster to represent sexual minorities in a more generalized manner within the sample. In academic qualifications, the categories “up to 9th grade”, “10th grade”, “11th grade”, “12th grade” were clustered into the category “12 years”.

Given some missing responses, proportional calculations were performed for the nominal variables, and the missing ages were assigned their mean value. Aiming to preserve the sample size, and since there were only a few missing responses involving age, we decided to use this method to maintain the statistical power of the sample, even though it may produce some biased estimates (Allison, 2002).

Descriptive statistics were then conducted to describe the sample: mean, standard deviation, maximum, minimum, mode, and median for age; and frequencies and percentages for the remaining sociodemographic variables. The gender variable’s “non-binary” category was presented for the sample description but excluded from the remaining analyses as it did not show a significant quantity. More descriptive statistics were performed to assess substance use (including excessive consumption), sexual violence, and psychological symptoms. The lack of frequencies and percentages of results of “hard drugs” resulted in the elimination of the items “cocaine or crack”, “heroin”, “methamphetamines”, “magic mushrooms (hallucinogens)”, “ecstasy/molly/MDMA”, and “other drugs” in the substance use variable.

For the levels of sexual violence, there were no significant results regarding the frequencies and percentages of perpetration items; so they were also excluded, thus only results for sexual victimization remained, which were used to create the new variable “sexual victimization (total)”. For the psychological symptom analysis, four new variables were created: three formed with each subscale’s mean results from the BSI-18 (“somatization”, “depression”, and “anxiety”), and one representing the overall mean score of the BSI-18 (“psychological symptoms (total)”).

Descriptive statistics were conducted for these new variables to obtain their means and standard deviations, aiming to compare these values with the reference values for community populations (Canavarro, 2007). Given the sample size, it is possible to assume a normal distribution based on the Central Limit Theorem, which posits that larger samples increase the likelihood that the distribution approximates normality.

Independent sample t tests were performed to verify whether dependent variables’ means (substance use and sexual victimization) differed significantly in the two comparison groups for each independent variable (men and women in gender, and heterosexual and non-heterosexual in sexual orientation). Pearson’s correlation tests were also performed to calculate the degree of association between substance use and sexual victimization, substance use and psychological symptoms, and sexual victimization and psychological symptoms; for this purpose, a new variable of “total sexual victimization” was created with the mean responses to items related to victims. Finally, Linear Regression Analysis was conducted to examine the predictive effects of sociodemographic variables and substance use on sexual victimization, and the predictive effects of substance use and sexual victimization on “total psychological symptoms”.

3. Results

For all statistical procedures, p < 0.05 was used. Since the questionnaire on substance use was built according to criteria that allowed for assessing the frequency of this behavior, each item was treated as a separate variable. As reported in

Table 2. Frequency of substance use per week.

	Never		1 Time		4/5 Times		6/7 Times
	n	%	n	%	n	%	n	%
Cigarettes or other tobacco	352	76.7	30	6.6	22	4.7	55	12
Electronic cigarette or vaping	418	91	29	6.3	5	1.1	7	1.6
Alcoholic beverages (beer, wine, etc.)	228	49.8	171	37.2	34	7.4	26	5.6
Alcoholic beverages (shots, spirits, etc.)	328	71.4	109	23.7	13	2.8	9	2
Hashish/Marijuana	428	93.2	21	4.5	5	1.1	5	1.1
Unprescribed medication	437	95.2	19	4.1	2	0.5	1	0.2

, consumption percentages are relatively low.

Alcoholic beverages such as beer or wine, were the most frequently used substances (50.2%). The remaining substances showed percentages of non-consumption above 70%. Only 8.8% of participants considered having consumed excessively (

Table 3. Perceived excessive consumption.

	n	%
Yes	40	8.8
No	419	91.2

).

Similar to substance use, sexual victimization frequency was assessed item by item, given the same nature of the questionnaire. Since there were no significant percentages, items related to the perpetrator were excluded. As for the items related to lifetime experiences of sexual victimization, the highest percentages were recorded for the questions “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (39.2%), “Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?” (37.1%) and “Has anyone ever kissed or touched you sexually without your consent?” (33.9%). The remaining items showed percentages under 18% (

Table 4. Frequency of sexual victimization experiences.

	No		Yes
	n	%	n	%
Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?	376	82	83	18
Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?	419	91.3	40	8.7
Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?	395	86	64	14
Has anyone ever kissed or touched you sexually without your consent?	303	66.1	156	33.9
Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?	279	60.8	180	39.2
Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?	289	62.9	170	37.1

).

As for psychological symptoms, the new variables were used to assess mean and standard deviation values. The results for “somatization” (M = 0.77; SD = 0.67), “depression” (M = 1.2; SD = 0.82), “anxiety” (M = 1.26; SD = 0.74) and “psychological symptoms (total)” (M = 1.08; SD = 0.66) are reported in

Table 5. Descriptive statistics for psychological symptom levels.

	Minimum	Maximum	Mean	Standard Deviation
Somatization	0.00	3.33	0.77	0.67
Depression	0.00	4.00	1.2	0.82
Anxiety	0.00	3.67	1.26	0.74
Psychological Symptoms (total)	0.00	3.61	1.08	0.66

. The minimum mean value was 0.00 for the four variables, and the maximum mean values were 3.33 for “somatization”, 4 for “depression”, 3.67 for “anxiety”, and 3.61 for “psychological symptoms (total)”.

Independent sample t test delivered the existence of statistically significant differences between men and women for levels of use of the substance “Electronic cigarette or vaping” (t (436) = −2.037; p = 0.042), with men presenting higher values (M = 1.26; SD = 0.78) than women (M = 1.13; SD = 0.56); and the substance “Alcoholic beverages (beer, wine, etc.)” (t (319) = −4.329; p < 0.001), with men also presenting higher values (M = 2.26; SD = 1.4) than women (M = 1.66; SD = 0.97) (

Table 6. Differences between substance use levels in men and women.

		M	SD	t(df)	Cohen’s d	p
Cigarettes or other tobacco	Women	1.65	1.36	−1.064(419)	−0.11	0.288
	Men	1.81	1.52
Electronic cigarette or vaping	Women	1.13	0.56	−2.037(436)	−0.21	0.042 *
	Men	1.26	0.78
Alcoholic beverages (beer, wine, etc.)	Women	1.66	0.97	−4.329(319)	−0.54	0.000 **
	Men	2.26	1.4
Alcoholic beverages (shots, spirits, etc.)	Women	1.36	0.77	−1.726(387)	−0.20	0.085
	Men	1.52	0.92
Hashish/Marijuana	Women	1.12	0.56	0.517(435)	0.06	0.605
	Men	1.09	0.44
Unprescribed medication	Women	1.06	0.31	0.088(436)	0.00	0.930
	Men	1.06	0.4

* p < 0.05. ** p < 0.001.

).

For sexual orientation, the test shows statistically significant differences between heterosexuals and non-heterosexuals for levels of the use of the substance “Hashish/Marijuana” (t (436) = −2.335; p = 0.02), with non-heterosexuals showing higher values (M = 1.25; SD = 0.82) than heterosexuals (M = 1.09; SD = 0.47). It is also possible to observe higher consumption levels of most substances by non-heterosexual participants; however, said levels cannot be considered statistically significant (

Table 7. Differences between substance use levels in heterosexuals and non-heterosexuals.

		M	SD	t(df)	Cohen’s d	p
Cigarettes or other tobacco	Heterosexual	1.63	1.36	−1.758(420)	−0.22	0.08
	Non-heterosexual	1.93	1.51
Electronic cigarette or vaping	Heterosexual	1.17	0.65	0.695(437)	0.08	0.487
	Non-heterosexual	1.12	0.55
Alcoholic beverages (beer, wine, etc.)	Heterosexual	1.76	1.08	−1.558(318)	−0.21	0.12
	Non-heterosexual	2.00	1.25
Alcoholic beverages (shots, spirits, etc.)	Heterosexual	1.38	0.78	−1.025(387)	−0.14	0.306
	Non-heterosexual	1.49	0.92
Hashish/Marijuana	Heterosexual	1.09	0.47	−2.335(436)	−0.29	0.02 *
	Non-heterosexual	1.25	0.82
Unprescribed medication	Heterosexual	1.05	0.28	−1.752(437)	−0.21	0.081
	Non-heterosexual	1.12	0.5

* p < 0.05.

).

Regarding the levels of sexual victimization, statistically significant differences were also found between men and women for the item “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (t (445) = 3.113; p = 0.002), with women presenting higher values (M = 2.64; SD = 2.39) than men (M = 1.88; SD = 1.89) (

Table 8. Differences between sexual victimization levels in men and women.

		M	SD	t(df)	Cohen’s d	p
Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?	Women	1.59	1.48	1.921(445)	0.21	0.055
	Men	1.31	0.99
Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?	Women	1.31	1.17	1.661(444)	0.17	0.097
	Men	1.13	0.59
Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?	Women	1.42	1.17	1.921(444)	0.21	0.055
	Men	1.19	0.87
Has anyone ever kissed or touched you sexually without your consent?	Women	2.05	1.84	1.73(445)	0.18	0.084
	Men	1.73	1.47
Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?	Women	2.64	2.39	3.113(445)	0.34	0.002 *
	Men	1.88	1.89
Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?	Women	2.21	2.02	1.158(444)	0.12	0.247
	Men	1.97	1.66

* p < 0.05.

).

As observed in

Table 9. Differences between sexual victimization levels in heterosexuals and non-heterosexuals.

		M	SD	t(df)	Cohen’s d	p
Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?	Heterosexual	1.34	1.06	−6.066(446)	−0.71	0.000 **
	Non-heterosexual	2.29	2.13
Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?	Heterosexual	1.17	0.79	−4.351(445)	−0.52	0.000 **
	Non-heterosexual	1.72	1.76
Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?	Heterosexual	1.23	0.8	−5.773(445)	−0.68	0.000 **
	Non-heterosexual	1.98	1.86
Has anyone ever kissed or touched you sexually without your consent?	Heterosexual	1.74	1.52	−5.777(446)	−0.68	0.000 **
	Non-heterosexual	2.9	2.31
Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?	Heterosexual	2.13	2.03	−6.093(446)	−0.71	0.000 **
	Non-heterosexual	3.71	2.82
Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?	Heterosexual	1.97	1.78	−4.138(445)	−0.48	0.000 **
	Non-heterosexual	2.9	2.4

** p < 0.001.

, regarding sexual victimization, all items show statistically significant differences between heterosexuals and non-heterosexuals (p < 0.001), and for all items, non-heterosexuals show higher values of victimization experiences, as the mean values for non-heterosexuals are higher than those for heterosexuals.

Regarding the levels of psychological symptoms, there were statistically significant gender-specific differences in the items “somatization” (t (453) = 2.917; p = 0.004), with women displaying higher levels (M = 0.83; SD = 0.68) than men (M = 0.63; SD = 0.61); “anxiety” (t (453) = 4.606; p < 0.001), with women also presenting higher levels (M = 1.36; SD = 0.76) than men (M = 1.01; SD = 0.66); and “psychological symptoms (total)” (t (453) = 2.805; p = 0.005), with women presenting higher values as well (M = 1.13; SD = 0.67) compared to men (M = 0.94; SD = 0.61) (

Table 10. Differences between psychological symptom levels in men and women.

		M	SD	t(df)	Cohen’s d	p
Somatization	Women	0.83	0.68	2.917(453)	0.30	0.004 *
	Men	0.63	0.61
Depression	Women	1.19	0.83	0.22(453)	0.01	0.826
	Men	1.18	0.8
Anxiety	Women	1.36	0.76	4.606(453)	0.48	0.000 **
	Men	1.01	0.66
Psychological Symptoms (total)	Women	1.13	0.67	2.805(453)	0.29	0.005 *
	Men	0.94	0.61

* p < 0.05. ** p < 0.001.

).

As for sexual orientation, it is possible to observe in

Table 11. Differences between psychological symptom levels in heterosexuals and non-heterosexuals.

		M	SD	t(df)	Cohen’s d	p
Somatization	Heterosexual	0.68	0.59	−6.301(454)	−0.73	0.000 **
	Non-heterosexual	1.15	0.81
Depression	Heterosexual	1.08	0.74	−6.128(454)	−0.71	0.000 **
	Non-heterosexual	1.64	0.96
Anxiety	Heterosexual	1.16	0.7	−5.779(454)	−0.68	0.000 **
	Non-heterosexual	1.65	0.81
Psychological Symptoms (total)	Heterosexual	0.97	0.58	−6.986(454)	−0.82	0.000 **
	Non-heterosexual	1.48	0.76

** p < 0.001.

that statistically significant differences were found for all items between heterosexuals and non-heterosexuals (p < 0.001). Non-heterosexual participants showed higher levels of psychological symptoms in comparison to heterosexuals.

Bivariate correlation tests were performed to compare the consumed substances and the experiences of sexual victimization. The Pearson correlation coefficients obtained showed significant correlations between “Cigarettes or other tobacco” and the items “Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?” (r = 0.221, p < 0.001), “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?” (r = 0.196, p < 0.001), “Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?” (r = 0.182, p < 0.001), and “Has anyone ever kissed or touched you sexually without your consent?” (r = 0.189, p < 0.001).

All of the former correlations were positive, meaning the higher the consumption of “Cigarettes or other tobacco”, the greater the probability of experiencing the sexual victimization types mentioned above. There were also significant negative correlations found between “Electronic cigarette or vaping” and “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (r = −0.099, p = 0.039), suggesting that the more “electronic cigarette or vaping” is consumed, the lower the probability of receiving virtual harassment. However, although the correlation is significant, it is quite weak.

The Pearson correlation coefficients obtained for the substance “Alcoholic beverages (beer, wine, etc.)” indicate positive significant correlations with the items “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist? (r = 0.113, p = 0.044) and “Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?” (r = 0.124, p = 0.027), meaning that the consumption of beer, wine, and similar alcoholic beverages is related to a higher probability of having unwanted sexual experiences under the influence and involving penetration.

For the substance “Alcoholic beverages (shots, spirits, etc.)” positive significant correlations were found for “Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?” (r = 0.125, p = 0.014), “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?” (r = 0.151, p = 0.003), “Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, or anal) without your consent?” (r = 0.124, p = 0.014), “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (r = 0.154, p = 0.002) and “Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?” (r = 0.13, p = 0.011). Hence, the more alcoholic beverages like shots and spirits are consumed, the higher the risk of experiencing the types of sexual victimization above.

Positive significant correlations were also found between “Hashish/Marijuana” and “Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?” (r = 0.192, p < 0.001), “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?” (r = 0.253, p < 0.001), “Has anyone ever initiated sexual contact with you, involving penetration (oral, vaginal, anal) without your consent?” (r = 0.239, p < 0.001), “Has anyone ever kissed or touched you sexually without your consent?” (r = 0.206, p < 0.001), and “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (r = 0.115, p = 0.016). Thus, a higher use of “Hashish/Marijuana” is correlated with more experiences of all sexual victimization types except verbal harassment.

“Unprescribed medication” also showed significant correlations with “Have you ever had sexual intercourse against your will, knowing the other person (friend, boyfriend/girlfriend, acquaintance, etc.)?” (r = 0.151, p = 0.002), “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?” (r = 0.208, p < 0.001), “Has anyone ever kissed or touched you sexually without your consent?” (r = 0.172, p < 0.001), “Has anyone ever sent you an email, SMS, message on a social network, phone call with sexual content, comments, jokes, stories, pictures, or videos that made you feel uncomfortable or offended?” (r = 0.146, p = 0.002) and “Has anyone ever made sexual comments about you or told jokes or stories of a sexual nature that you found uncomfortable or offensive?” (r = 0.116, p = 0.015). Since these correlations are all positive, the more “Unprescribed medication” is taken, the more the occurrences of victimization described above will be experienced. Correlational studies were performed for the new variable “Sexual victimization (total)” as well, resulting in significant positive correlations with the substances “Cigarettes or other tobacco” (r = 0.183, p < 0.001), “Alcoholic beverages (shots, spirits, etc.)” (r = 0.163, p = 0.001), “Hashish/Marijuana” (r = 0.204, p < 0.001) and “Unprescribed medication” (r = 0.108, p < 0.001), meaning the more those substance are consumed, the higher the risk of going through general sexual victimization experiences.

Bivariate correlation tests were also performed to compare the consumed substances and psychological symptoms. The Pearson correlation coefficients obtained showed significant correlations between somatization and “Cigarettes or other tobacco” (r = 0.158, p = 0.001), “Alcoholic beverages (beer, wine, etc.)” (r = 0.125, p = 0.025), and “Alcoholic beverages (shots, spirits, etc.)” (r = 0.107, p = 0.034); depression and “Cigarettes or other tobacco” (r = 0.12, p = 0.013) and “Alcoholic beverages (beer, wine, etc.)” (r = 0.154, p = 0.006); anxiety and “Cigarettes or other tobacco” (r = 0.1, p = 0.04), and “Unprescribed medication” (r = 0.102, p = 0.033); and total psychological symptoms and “Cigarettes or other tobacco” (r = 0.141, p = 0.003), “Alcoholic beverages (beer, wine, etc.)” (r = 0.127, p = 0.023), and “Unprescribed medication” (r = 0.104, p = 0.028). All correlations found were positive, which means that higher levels of psychological symptoms are correlated with higher levels of said substance consumption (Table S1).

To assess the impact of both SU and SV on the total of psychological symptoms, three models were performed using Linear Regression. Model 1 evaluated only sociodemographic characteristics as possible predictors. In Model 2, in addition to sociodemographic characteristics, the consumption of each substance was introduced. Finally, Model 3 encompassed sociodemographic characteristics, substance use, and the total sexual victimization. These can be found in

Table 12. Linear Regression Analysis predicting psychological symptoms—Models 1, 2 and 3.

	Model 1			Model 2			Model 3
	B	Error	β	B	Error	β	B	Error	β
Age	−0.034	0.015	−0.148 *	−0.032	0.015	−0.137 *	−0.032	0.014	−0.139 *
Gender	−0.171	0.090	−0.115	−0.228	0.093	−0.154 *	−0.209	0.089	−0.14 *
Citizenship	0.129	0.113	0.073	0.113	0.115	0.064	0.107	0.109	0.060
Marital Status	−0.005	0.017	−0.017	0.000	0.017	0.001	−0.007	0.017	−0.025
Academic Qualifications	0.066	0.069	0.061	0.075	0.069	0.070	0.087	0.066	0.080
Place of Residence	0.068	0.040	0.107	0.059	0.040	0.093	0.035	0.038	0.056
Socio-Economic Status	0.033	0.062	0.032	0.019	0.062	0.018	0.056	0.059	0.056
Sexual Orientation	0.403	0.099	0.247 **	0.372	0.101	0.228 **	0.230	0.100	0.141 *
Cigarettes or other tobacco				0.070	0.041	0.126	0.053	0.039	0.095
Electronic cigarette or vaping				−0.014	0.072	−0.012	0.012	0.069	0.011
Alcoholic beverages (beer, wine, etc.)				0.033	0.061	0.046	0.027	0.058	0.038
Alcoholic beverages (shots, spirits, etc.)				0.062	0.072	0.069	0.028	0.069	0.031
Hashish/Marijuana				−0.113	0.081	−0.094	−0.165	0.077	−0.137 *
Unprescribed medication				0.120	0.100	0.073	0.012	0.097	0.008
Sexual victimization							0.181	0.035	0.335 **
R2			0.084			0.094			0.181
F			3.951			2.923			4.826

* p < 0.05. ** p < 0.001.

.

Model 1 was significant (F(8; 251) = 3.951; p < 0.001), explaining 8.4% of the variance in psychological symptoms levels in the sample. Age (p = 0.02) and sexual orientation (p < 0.001) emerged as significant predictors. Thus, the older a person is, the fewer psychological symptoms they will report, and non-heterosexuals tend to report more symptoms than heterosexuals.

Model 2 was also significant (F(14; 245) = 2.923; p < 0.001); when introducing each substance’s use, the model explains 1% more PS levels than the previous model3. Here, the significant predictors continue to be the variables age (p = 0.032) and sexual orientation (p < 0.001), from which the same conclusions can be drawn. However, gender emerges as well (p = 0.015), according to which men are more likely to report symptoms.

Model 3, which was significant (F(15; 244) = 4.826; p < 0.001) like the previous models, explained 18.1% of the variance with the introduction of sexual victimization, which was considered a significant increase compared to Models 1 and 2. In this model, there are five predictors of psychological symptoms. Age (p = 0.022), gender (p = 0.019), and sexual orientation (p = 0.022) are still predictors, from which the same conclusions mentioned in Models 1 and 2 can be drawn. In addition, the new predictors are “Hashish/Marijuana” (p = 0.034), in which the more this substance is consumed, the less the symptoms are felt; and sexual victimization (p < 0.001), with greater symptoms being reported with more of these experiences. Sexual victimization is the predictor that has the most impact on psychological symptoms.

4. Discussion

The main purpose of this study was to assess substance use, sexual violence, and psychological symptom levels in a community sample of Portuguese college students. The results show that psychological symptom levels in this sample of college students presented higher values when compared to reference values for the community sample (Canavarro, 2007). This supports the information presented in the beginning of this study reporting that college students suffer more from mental health issues and can be explained by challenges that come in this moment of life when young adults start their independent life away from parents but also worry about their academic workload and performance (Cao et al., 2020; Erschens et al., 2019; Ishii et al., 2018; Jenkins et al., 2019; Kecojevic et al., 2020; Maia & Dias, 2020; McDougall et al., 2019; Park et al., 2020; Ratanasiripong et al., 2018; Rosenthal et al., 2018; Trigueros et al., 2020; Usher & Curran, 2019; Wang & Zhao, 2020).

Women showed significantly higher levels of somatization, anxiety, and psychological symptoms in general compared to men, and even though not significant, the same can be said about depression. Additionally, non-heterosexuals also showed very significantly higher levels of symptoms in general, as well as levels of each symptom. These two results are consistent with the literature considering Meyer’s Minority Stress Theory (Meyer, 2003), which explains how minority groups, especially SGMs, tend to have worse mental health due to all the difficulties related to stigma and prejudice.

Concerning substance use, results show that alcoholic beverages such as beer or wine are the substances university students tend to use most frequently, whereas the remaining substances show a high percentage of non-consumption. These results show how common the correlation between academic social life and alcohol consumption is, especially with cheaper drinks such as wine and beer, since alcohol seems to be an instrument for socialization, whether during academic festivities or simply in social interactions (Santiago, 2021).

There were also significant differences in consumption regarding gender, with men stating that they used or consumed alcoholic beverages such as beer or wine and electronic cigarettes or vaping more than women, which is consistent with the literature (Blanco et al., 2018; Colby et al., 2009; SICAD, 2022; Alves, 2024). Even if not significant, men also mentioned that they consumed more cigarettes or other tobacco and alcoholic beverages such as shots and spirits, which is consistent with the literature; women reported that they consumed more cannabis products (hashish or marijuana), which contrasts with the previous literature where men are typically the primary consumers; and there were no significant differences in gender for unprescribed medication use, different from previews research that shows that women tend to self-medicate more (Blanco et al., 2018; SICAD, 2022).

Furthermore, in general, non-heterosexuals reported more consumption than heterosexuals, although only cannabis products had significant results, which is also congruent with the literature, considering the possibility of substance use as a coping mechanism to deal with stigma, discrimination, and past experiences of SV (Chaiton et al., 2021; Salerno et al., 2021; Azagba et al., 2022; Marcantonio & Willis, 2022).

The use of substances, licit or illicit, is frequently related to poorer mental health. However, it is unclear how exactly this correlates with several studies mentioning different types of “drugs” having an impact on different psychological symptoms (Blanco et al., 2018; Walters et al., 2018). In this study, we found positive correlations among some substances and psychological symptoms, meaning that higher levels of psychological symptoms are correlated with higher levels of said substance consumption: “Cigarettes or other tobacco” was correlated with all symptoms; “Alcoholic beverages (beer, wine, etc.)” was correlated with all symptoms except for anxiety; “Alcoholic beverages (shots, spirits, etc.)” were correlated solely with somatization, possibly in relation to hangovers; and “Unprescribed medication” correlated with anxiety and total psychological symptoms.

However, when assessing the impact of substance use on the total psychological symptoms, it was found that when more hashish/marijuana is consumed, the fewer symptoms are felt. This result was unexpected. However, the continuous consumption of this substance may reduce affective responses such as positive affect, stress, and anxiety (DeAngelis & al’Absi, 2020), which may have contributed to affective blunting in the participants.

Results for sexual violence showed that the experiences with higher frequency were those related to online behavior, sexual comments, and forced kisses or touches. Compared to the remaining items, these seem to be the ones describing less invasive forms of sexual violence. Significant differences were found between men and women relating to online behavior, specifically, with women reporting having received more undesirable sexual content than men. These forms of sexual violence with less contact are probably more common in this sample since it is composed mostly of young adults who are intimately connected to technology, the internet, and social media, which has become a means to spread unwanted sexual content over the years (Associação de Mulheres Contra a Violência [AMCV], 2023; Reed et al., 2020), possibly due to the anonymity of the perpetrator who will not understand or feel the consequences of their actions. The remaining “less invasive” form of sexual aggression is probably more frequent since socially and criminally they do not hold the same “legal weight” as a rape, as observed in the high number of reported cases in Portugal (Associação Portuguesa de Apoio à Vítima [APAV], 2023). Even though the remaining differences are not significant, it is worth mentioning the higher levels of experiences in women, as reported in the existing literature (Coulter et al., 2017; Gómez, 2022; Hines et al., 2012; Ortensi & Farina, 2020).

All unwanted sexual experiences described are significantly higher in non-heterosexuals. This is also consistent with previous research indicating that LGBTQIA+ individuals are more at risk of suffering sexual victimization due to discrimination and stigma (Coulter et al., 2017; Martin-Storey et al., 2018; Meyer, 2003; Ortensi & Farina, 2020; Snyder et al., 2018; Marcantonio & Willis, 2022). It was observed that going through experiences of sexual victimization has a negative impact on mental health, as all items and the variable for total sexual victimization were strongly and positively correlated with psychological symptoms.

When assessing the impact of sexual victimization on psychological symptoms, it was observed that SV has the most impact on mental health, with a higher number of experiences associated with more frequent or more intense symptoms. These results are also consistent with the literature, where anxiety and depression symptoms and disorders are highly correlated with having experienced SV, since these are intense and traumatic situations (Binion & Gray, 2020; Depraetere et al., 2023; Gómez, 2022; Tarzia et al., 2020; Astle et al., 2023).

When assessing correlations between sexual victimization and substance use, correlations were found between different drugs and different experiences of sexual violence, the most prominent being the sexual victimization experience “Have you ever had sexual intercourse against your will under the influence of alcohol or drugs, because you didn’t have the strength to resist?” and the most prominent substances “Alcoholic beverages (shots, spirits, etc.)”, “Hashish/Marijuana”, and “Unprescribed medication”. Although survivors are never to blame for suffering SV under any circumstances (Basile et al., 2021), this correlation may happen in two ways: (1) the consumption of these particular substances has a depressant or hallucinogenic effect, therefore leaving the victim unconscious or unable to resist and more prone to be taken advantage of as stated in the sentence; or (2) these “drugs” are used as a way to cope with past sexual violence experiences (e.g., Lorenz & Ullman, 2016)

College students are usually more involved in dating and new technologies, and even though the most reported victimization experience involves unwanted sexual behavior online, a lot of times, sexual violence is not even perceived in these contexts due to deep-rooted beliefs and stereotypes that keep people from understanding the severity of abusive behaviors in relationships and through social networks. It is also worth mentioning the impact COVID-19 has most likely had on sexual violence, since more cases of domestic violence emerged due to the forced permanent coexistence that came with the lockdowns (Ribeiro et al., 2022); substance use, since this escalated as a coping mechanism to deal with stress caused by the pandemic; and psychological symptoms, since lockdowns forced isolation that led to feelings of loneliness and depression (Banerjee & Rai, 2020), and the anxiety and somatization that came with the pandemic (Ho & Ho, 2020; Shevlin et al., 2020).

This study provides evidence that substance use and sexual violence have negative effects on mental health, with severe effects among individuals from minority groups such as SGMs, despite some authors identifying limitations in the SMT, such as the lack of studies that associate genetic factors, personality traits, and PSs (e.g., Michael Bailey, 2020).

In addition to its contributions, it is important to point out certain limitations. For instance, since data collection was carried out through a survey, voluntary bias and social desirability bias may presumably bias the results. Considering that the sample was collected through convenience sampling, an unwilled selection of certain participants with specific characteristics was held (e.g., having access to a computer or the internet), building a barrier to a range that would allow us to achieve a more balanced sample. The demographic profile was also unbalanced, as the majority of participants identified as women and heterosexual, not allowing for the possibility of analyzing non-binary individuals.

Regarding methodological components, there is a concern with the use of imputation for age. Although it was applied minimally and without excluding data, it still introduces potential bias. Concerning correlations, although weak correlations are common in social science studies, they may limit our ability to understand the true dependence between the studied variables, highlighting the need for further research with larger samples. Even so, the lack of longitudinal data prevents us from understanding how these individuals will evolve over time, which could be addressed in future studies, and prevents causal relationships from being established.

Additionally, themes regarding sexual violence are often sensitive subjects to approach. Considering this, and similar to what occurred in the study by Martin-Storey et al. (2018), where victims of sexual violence may have approached the survey in two opposite ways: (1) some may have been drawn to the survey given their experience, and (2) others may have avoided the survey as to avoid discomfort, feelings of shame, or triggering questions associated with SV trauma. Thus, these approaches may have interfered with the results. From a different angle, possibly out of fear of repercussions, even with the assurance of confidentiality and anonymity, there was a shortage of participants with perpetrating or abusive actions, which made statistical evaluations unfeasible and led to their exclusion from the study.

This study offers a significant contribution to the understanding of the current state of psychological symptoms, substance use, and experiences of sexual violence among Portuguese university students. To our knowledge, it is the first study in Portugal to examine these variables in an integrated manner.

It is recommended that future qualitative research explore psychosymptomatology, experiences of sexual violence, substance use, and trauma-related symptoms such as PTSD. Such research could provide valuable insights to guide interventions and inform the development of social, educational, political, and health policies.

5. Conclusions

Sexual violence and substance use are shown to negatively affect the mental health of college students, promoting the development or the escalation of existing psychological symptoms, especially when it comes to sexual and gender minorities. These findings contribute not only to the existing literature on this topic, especially in Portugal, but also emphasize the importance and need for more inclusive social, educational, political, and health policy-related studies.

Implementing educational and preventive projects in universities that allow the restructuring of beliefs and values that may legitimize attitudes of sexual violence and explore the notion of consent through methods that combat victim blaming, emphasize responsibility as adults (Pires et al., 2018; Girod, 2024), and promote the empowerment of individuals who are more vulnerable and at risk of suffering sexual violence is necessary.

In terms of substance use, continuing and expanding substance use prevention projects, such as the “Risks & Challenges Program” ([Programa Riscos & Desafios] Rocha et al., 2018), are necessary but they could be expanded across the entire undergraduate, master’s, and doctoral program cycle. It is crucial to talk about the main substances consumed and explain their effects on the body and mind, the risks involved in self-control, conscious consumption, and combating peer pressure.

To discourage potential aggressors, informal education campaigns should be encouraged to emphasize that consent is much more than a lack of response (Pires et al., 2018; Girod, 2024) and that substance use does not excuse sexual violence (Ison et al., 2025). Furthermore, greater efforts are needed to raise awareness among staff at academic parties, enabling them to identify harassment situations, to take responsibility for the spaces where violence occurs (e.g., nightclubs, bars, among others), as well as to raise awareness among staff to identify harassment and violence situations and combat communication surrounding this issue (Pires et al., 2018).

At the clinical and educational levels, it is suggested that Portuguese universities should contract more mental health care providers on campus. Specifically, this group of professionals should invest in training and specialized topics related to minority groups (Neves et al., 2023), as well as in the promotion of trauma-informed classrooms (Sweetman, 2022), to understand their particularities, intervene effectively, promote safe spaces, include more counselling and behavioral support personnel, and raise awareness on campus.