Singlehood–Mental Health Associations in Sexual and Gender Minority Youth Assigned Female at Birth: A Longitudinal Study

Abstract

This longitudinal study examined the relationships between multiple dimensions of singlehood and mental health, along with their moderators, in a diverse cohort of sexual and gender minority youth assigned female at birth (N=205, M_age=19.5; 39% <18 years old; 27% gender minority; 74% racial/ethnic minority) who were single (no romantic partnership) at least once across 7 visits (retention >90% at all visits). Singlehood dimensions included single status, length of singlehood, and transitions into and out of singlehood at each visit, as well as proportion of singlehood periods across all visits. At the between-person level, individuals with a higher proportion of singlehood periods across the study reported higher stress, anxiety, and depression. At the within-person level, while youth experienced increased depression when single, their depression and stress decreased as the length of singlehood increased. Additionally, fear of being single (FOBS) and perceived social support, each showing unique associations with mental health outcomes in this study, were included as moderators. Notably, only those with low to average (not high) FOBS experienced a reduction in mental health problems as singlehood length increased. Taken together, the findings shed light on the factors associated with mental health and well-being in single individuals from a vulnerable and marginalized population.

Sexual and gender minority (SGM) individuals, including lesbian, gay, bisexual, transgender, and/or non-binary individuals, experience higher rates of mental health problems (e.g., depression) than their cisgender and heterosexual counterparts (Amos et al., 2020; Lytle et al., 2014; Marshal et al., 2013; Moore et al., 2021). Within the SGM community, disparities in mental health problems are largest for youth and emerging adults assigned female at birth (AFAB; Lipson et al., 2019; Marshal et al., 2008, 2013). Relationship involvement is a key protective factor for mental health in SGM-AFAB youth, such that partnered individuals report fewer mental health problems than their single peers (Whitton et al., 2020, 2021).

However, this does not mean that being single is inherently bad for mental health. Many SGM and non-SGM singles do not suffer (and even thrive) from being single (Baams et al., 2014; Willitts et al., 2004). Despite the multidimensional and complex nature of the issue, relationship science often excludes singles from the discussion of well-being, unless they function as a point of comparison for their partnered peers (Girme, Park, et al., 2022). This between-group comparison ignores the within-group variability among singles that can point to the factors associated with their well-being (Girme, Park, et al., 2022). In addition, relationship science has often ignored the unique experiences of singlehood for SGM youth who experience barriers to relationship formation. These include limited pool of partners and the constant risks of stigma, discrimination, and rejection upon disclosing their SGM identity to romantic interests (DeHaan et al., 2013; Korchmaros et al., 2015; Owens et al., 2023).

This study explores the associations between singlehood dimensions (beyond current status) and mental health in SGM-AFAB youth, the population already facing significant mental health disparities. Our objectives are twofold: 1) to identify the associations and 2) to better understand these associations through contextual factors closely linked to singlehood, namely fear of being single and social support.

Singlehood and Mental Health

Research on the general population has documented associations between singlehood status–defined as no romantic partnership–and poor mental health. Single women tend to report lower psychological well-being and more mental health problems than romantically partnered women (Kamp Dush & Amato, 2005; Simon & Barrett, 2010; Willitts et al., 2004). However, the theoretical frameworks of these studies have been rooted in heteronormative narratives that uphold conventional norms (e.g., marriage, family) and depict singles as deficient or deviant (for a review, see Girme, Park, et al., 2022). It therefore cannot be assumed that these associations will generalize to SGM individuals, whose lives and relationships operate outside of heteronormativity. On the other hand, the associations between singlehood and poor mental health may be exacerbated for SGM individuals because people who are queer and single could experience stigma due to each of their marginalized identities. Though relatively few in number, studies exploring these questions with SGM-AFAB samples have found that, like in cisgender heterosexual samples, singlehood status is associated with worse mental health (Ayala & Coleman, 2000; Kornblith et al., 2016; Whitton et al., 2018, 2020, 2021).

The experience of singlehood among SGM individuals is likely unique and shaped by minority stress related to their SGM status (Meyer, 1995). Many SGM people remain single due to a collective history of stigma and discrimination (Hostetler, 2009), and those with higher levels of internalized SGM stigma are more likely to remain single for longer periods (Laming et al., 2023). Additionally, many same-sex relationships have assimilated through monogamy, gender conformity, and domesticity (Lavender-Stott, 2023). Those who align with conventional family values find acceptance, while deviations from heteronormative relationship norms (e.g., being single or having only sexual partners) result in further marginalization (Lavender-Stott, 2023). Consequently, single SGM individuals may find themselves navigating a dual challenge: coping with the minority stress that contributes to their singlehood while also facing societal pressures to adhere to conventional relationship norms. This intersectionality highlights the need for a nuanced approach in researching the link between singlehood and mental health among SGM populations.

Existing research predominantly centers on the between-person effect of singlehood status (single vs. not) on mental health, using cross-sectional designs. Much less understood are the multiple dimensions of singlehood (e.g., singlehood length or transitions) that can fluctuate over time and may explain their mental health above and beyond the ‘state’ of being single. First, it is unclear how the length of singlehood may affect psychological well-being. Though extended periods of being single might allow individuals the opportunity to enjoy the benefits associated with single life (e.g., increased freedom), it is also possible that they may worsen mental health due to prolonged romantic loneliness. Research on the relationship between the length/duration of singlehood and mental health, especially among SGM youth, is very limited. Most studies focus on long-term singlehood in (relatively) older adults, often without directly exploring its impact on mental health (Hostetler, 2012; Laming et al., 2023; Timonen & Doyle, 2014). Further, conflicting findings have emerged: one cross-sectional study (Schachner et al., 2008) shows that long-term single adults (i.e., not being in a relationship for at least three years and not anticipating one in the near future) reported higher levels of mental health problems compared to their partnered counterparts. However, this study only compares between-group statuses and does not consider the progressive duration of singlehood. In contrast, a longitudinal study shows that adults tend to become more content with singlehood as they age (Park et al., 2022), suggesting a possible adaptation effect.

While existing literature predominantly focuses on major life events, such as entering marriage, experiences relating to transitions into or out of singlehood receive much less attention. These latter transitions, which are more transient and may involve significant psychological adjustments for youth, are often overshadowed by more traditional milestones centered on romantic relationships (singular events of entering/exiting a relationship). The emotional and social adjustments required during periods of singlehood transitions can have profound impacts on mental health. For example, the period following a breakup may be marked by a significant decline in mental health, suggesting that the transition phase itself, rather than the state of being single, could be the critical factor influencing well-being. Understanding these dynamics is essential to disentangling the effects of relationship status from the effects of transitions, thereby preventing the misattribution of mental health declines to singlehood itself. Research has shown that a recent breakup is associated with poorer mental health (Simon & Barrett, 2010; Wade & Pevalin, 2004), though the temporality of this association is unclear. Conversely, transitioning into partnerships is linked to better well-being (Barr et al., 2016; Bühler et al., 2023), but this benefit is contingent on the quality of the new relationship (Barr et al., 2016). These findings underscore the need to look beyond relationship status and explore the distinct impacts of transitions themselves.

The current study offers a nuanced analysis of both transient and enduring aspects of singlehood and their distinct impacts on mental well-being. By leveraging multi-wave data, we address the limitations inherent in cross-sectional research. For example, instead of relying on retrospective recalls, our design captures the actual progression of singlehood length over time. In sum, this longitudinal approach allows us to explore not only the overall association between singlehood and mental health across individuals but also the variations within individuals over time.

Moderators: Fear of Being Single and Social Support

We also attempt to move the field beyond simple demonstrations of associations between singlehood and mental health by exploring socioemotional factors that may moderate these associations. In particular, we explore whether fear of being single and social support—factors that are theoretically and empirically linked to singlehood—buffer or exacerbate the singlehood-mental health association.

Fear of Being Single

Fear of being single (FOBS) entails “concern, anxiety, or distress regarding the current or prospective experience of being without a romantic partner” (Spielmann et al., 2013, p. 1050). While research has established the link between FOBS and poor well-being in non-SGM adults (Girme, Park, et al., 2022), its implications for SGM youth have yet to be explored. Given the common challenges of being single (e.g., loneliness), along with the added minority stress experienced by SGM individuals, it is crucial to examine whether FOBS exacerbates mental health challenges associated with singlehood in this population.

People often experience social pressure to have a relationship (Moorman, 2020; Sprecher & Felmlee, 2021) and stigma or negative stereotypes associated with being single (A. N. Fisher & Sakaluk, 2020; Girme, Sibley, et al., 2022; Greitemeyer, 2009; Hertel et al., 2007). SGM-AFAB youth in particular may fear being single partly due to unique barriers to romantic relationship formation (e.g., limited pool of partners). Their FOBS may not only reflect anxiety over the absence of romantic partners but also broader apprehension about the dual challenges of being single and SGM. Such fears could cultivate a negative perception of singlehood (e.g., viewing it not as a choice but as an unavoidable condition), which in turn exacerbate the typical challenges associated with being single.

In this study, we examine how negative emotions or perceptions related to being single help us better understand the link between singlehood and mental health. Specifically, we are interested in how individual differences in FOBS might moderate the associations between singlehood and mental health problems. For instance, although involuntary singlehood is linked to lengthier period of being single (Apostolou & Wang, 2019), many, especially those from marginalized groups, may choose prolonged (or even life-long) singlehood (DePaulo, 2023; Kislev, 2023; Lavender-Stott, 2023) for reasons of empowerment and personal growth (Beckmeyer & Jamison, 2023; Moorman, 2020; Timonen & Doyle, 2014). However, those high in FOBS may not view their singlehood as an intentional choice (Adamczyk et al., 2021; Beckmeyer & Jamison, 2023). It is possible that SGM youth with high FOBS experience greater mental health declines when single due to their negative perceptions and anxieties about being single. This FOBS could also hinder their ability to enjoy the mental health benefits of singlehood (e.g., increased autonomy) that often accompanies prolonged periods of being single.

Social Support

Being single is often associated with loneliness and lower perceived support, which are strongly linked to poorer well-being (Adamczyk, 2016; Adamczyk & Segrin, 2015; Girme, Sibley, et al., 2022). While romantic partnerships provide crucial emotional support (Hombrados-Mendieta et al., 2013) and can buffer SGM individuals from SGM-related stressors (Feinstein et al., 2016; Whitton et al., 2018), many singles maintain robust social networks that offer significant support. Single (vs. partnered) people tend to have larger social networks and better social support functioning with family, neighbors, and friends (Sarkisian & Gerstel, 2016), They are also more likely to prioritize, invest in, and benefit from these relationships (A. N. Fisher et al., 2021; McLaughlin et al., 2002; Park et al., 2021; Sarkisian & Gerstel, 2016). For SGM people, strong community ties, friend-centered networks, and ‘family of choice’ are crucial sources of social support that help reduce reliance on romantic partners (Holcomb et al., 2022; Lavender-Stott & Allen, 2023; Leahy & Chopik, 2020; Paceley et al., 2017). Many also engage in non-traditional relationships (e.g., queer platonic relationships, solo-polyamory) that fulfill intimacy needs usually reserved for romantic relationships in traditional heterosexual contexts (Lavender-Stott, 2023). These robust social support from families of choice and close relationships not only offers emotional and instrumental support but also caregiving (Thomeer et al., 2021). The strong emphasis on this extensive network might lessen loneliness in many single SGM individuals. Indeed, research suggests that it is the absence of friendship support, rather than singlehood itself, that is a key determinant of loneliness (Dykstra, 1995).

In this study, we are interested in exploring whether social support moderates the link between singlehood and mental health. If the reduction of social support is the primary mechanism through which singlehood adversely affects mental health, then individuals who maintain strong social connections should not experience the detrimental effects of singlehood on mental health to the same extent as those with lower social support.

Despite numerous studies on the mental health benefits of social support among SGM youth and adults (Jacmin-Park et al., 2022; Leahy & Chopik, 2020; McDonald, 2018; Snapp et al., 2015; Watson et al., 2019), little is known about how social support moderates the relationships between various singlehood dimensions and mental health in SGM-AFAB youth. One study shows that high levels of support from family and significant others can buffer the impact of prolonged singlehood on romantic loneliness (Adamczyk, 2016). Given the role of social support in reducing loneliness and improving well-being, it is crucial to investigate whether SGM youth with different levels of social support experience different mental health outcomes related to singlehood.

Current Study

This longitudinal study explored the association between singlehood and mental health in a diverse group of SGM-AFAB youth aged 16–29 (39% < 18 years old; 27% gender minority; 74% racial/ethnic minority). One aim of the study (Aim 1) was to identify how different dimensions of singlehood are linked to mental health. Using data from 7 visits over the span of 3.5 years, we examined the effects of several singlehood dimensions, including singlehood status (time-varying), length of singlehood (time-varying), transition into and out of singlehood (time-varying) and proportion of singlehood throughout the study period (static/average), on mental health problems (stress, depression, and anxiety). Another aim (Aim 2) was to identify moderators (FOBS and social support) of the associations between singlehood dimensions and mental health problems.

It is worth emphasizing that the current study is exploratory in nature. We approached our research questions using a sub-sample of pre-existing data from a larger longitudinal project designed to broadly examine health and relationships among SGM-AFAB youth. The primary intent of this study is to provide preliminary insights into the singlehood-mental health association in this population, given the notable lack of research in this area.

Method

Supplemental materials, including measures, syntax, power analysis, singlehood variables by study visit, and detailed examples for singlehood coding are available on OSF: https://osf.io/uqsvz/

Participants

Data were drawn from a larger longitudinal cohort study examining SGM-AFAB youth’s health and relationships in the U.S. All participants were AFAB, aged 16–20 years at initial enrollment in their cohort of origin¹, lived in the Chicago area, and identified as a sexual or gender minority or reported same-gender attractions/sexual behavior. Participants were recruited using venue-based recruitment (e.g., SGM community organizations, health fairs), social media advertisements, and incentivized snowball sampling ($10 for every successfully recruited peer). All participants completed a baseline visit with follow-up visits every six months (except visit 7, which occurred 12 months after visit 6) and were paid $50 for each assessment. All study protocols were approved by Northwestern University’s Institutional Review Board with a waiver of parental permission for minor participants, as the study posed minimal risk and requiring parental consent would deter many SGM adolescents from participating due to fear of being outed (C. B. Fisher et al., 2016; Macapagal et al., 2017).

All baseline visits took place in person in a university setting. For subsequent visits, participants had the option to complete assessments online before the study switched to fully remote in 2020 during the COVID-19 pandemic. Participants completed self-report measures of health behaviors, mental health, and psychosocial variables in line with the original project’s objectives.

The current study used data from 205 participants who reported being single in at least one visit. For descriptives of each singlehood variable (e.g., singlehood status) at each visit, see supplemental materials. Table 1 shows sample demographics at baseline.

Table 1.

Demographics at Baseline (N=205)

	n	%
Age (M=19.47, Mdn=18.62, SD=3.18, Range = 16–29)
Gender Identity
Woman	150	73.2%
Transgender man	21	10.2%
Gender non-conforming	12	5.9%
Genderqueer	11	5.4%
Non-binary	10	4.9%
Not listed	1	0.5%
Sexual Identity
Gay	7	3.4%
Lesbian	39	19.0%
Bisexual	69	33.7%
Queer	29	14.1%
Unsure/questioning	12	5.9%
Straight/heterosexual	3	1.5%
Pansexual	34	16.6%
Asexual	10	4.9%
Not listed	2	1.0%
Race/Ethnicity
Non-Hispanic Black	72	35.1%
Non-Hispanic White	54	26.3%
Hispanic/Latinx	43	21.0%
Asian	14	6.8%
Biracial/Multiracial	20	9.8%
Not listed	2	1.0%
Education
8th grade or lower	2	1%
Some high school	96	46.8%
GED or high school diploma	12	5.9%
Some college	87	42.4%
Undergrade or equivalent degree	4	2%
Some graduate school	2	1%
Graduate degree	2	1%
Annual Income
< $20,000	192	93.7%
$20,000 - $39,999	9	4.4%
$40,000 - $49,999	2	1.0%
$50,000 - $59,999	1	0.5%
Partnership at Baseline
Yes	54	26.3%
No	151	73.7%

Measures

Singlehood.

At each visit, participants were asked to report on up to three partnerships they had in the past 6 months. Those who reported any partnerships provided further details, including partnership type (ranging from one-time sexual encounters to long-term/committed relationships), current status (and if no longer ongoing, the date of breakup), and partnership duration. Those who reported only partnerships that were purely sexual (e.g., hookups) were considered ‘single’ in this study. For details about how we coded different dimensions of singlehood (e.g., status, transition, length), see the Data Analysis section (Level 1 Predictors and Level 2 Predictors).

Fear of being single.

At any visit (except for baseline) when participants reported being single, they completed a 7-item version of the FOBS Scale (Spielmann et al., 2013) assessing discomfort with their current singlehood status (e.g., “It scares me to think that there might not be anyone out there for me”) on a 5-point scale (1=not at all true, 5=very true). We modified the scale to make it more appropriate for SGM youth (e.g., dropping “I need to find a partner before I’m too old to have and raise children”; see supplementals for detail). Items were averaged to create a total score, for which higher values indicated stronger FOBS (ω Range=.90-.95).

Social support.

Using the 12-item Multidimensional Scale of Perceived Social Support (Zimet et al., 1988), participants reported levels of perceived social support from friends (e.g., “I have friends with whom I can share my joys and sorrows”), family (e.g., “I get the emotional help and support I need from my family”), and special persons (e.g., “There is a special person who is around when I am in need”) on a 7-point scale (1=very strongly disagree, 7=very strongly agree) at all waves. Notably, one might argue that the concept of “special persons” are not pertinent to single individuals. However, our choice to include these items was guided by their relevance to queer relationships, where the term can encompass non-romantic significant others (e.g., chosen family, platonic life partner; Lavender-Stott, 2023). It is worth highlighting that our conclusions remained the same even when excluding these items from the analysis. Items were averaged to create a total score, for which higher values indicated stronger social support (ω Range=.88-.92).

Stress.

At all waves, we measured general stress in the past month using the 4-item Perceived Stress Scale (Cohen, 1988; Cohen et al., 1983). Participants rated how frequently they experienced each stressed feeling (e.g., “felt difficulties were piling up so high that you could not overcome them”) on a 5-point scale (0=never, 4=very often). Items were summed to create a total score, for which higher values indicated higher stress (ω ranged from .63–.79 across visits).

Depression and anxiety.

At all waves, we assessed depressive and anxious symptoms in the past week using the Patient-Reported Outcomes Measurement Information System (PROMIS) Depression Short Form 8a and Anxiety Short Form 8a (Pilkonis et al., 2011). Participants rated how frequently they experienced each of 8 depressed feelings (e.g., unhappiness, helplessness) and 8 anxious feelings (e.g., unease) on a 5-point scale (1=never, 5=always). Items were summed to create a total score, for which higher values indicated more depressive and anxiety symptoms (ω ranged from .94–.96 across visits for depression and from .93–.96 for anxiety).

Data Analysis

We conducted multilevel modeling using MPlus and maximum likelihood robust (FIML) estimation. Assessment visits/occasions (Level 1) were nested within participants (Level 2). Dependent variables (depression, anxiety, stress) were predicted using random intercept and slope models, which allowed the initial values and slopes (across study visits) to vary between persons.

Level 1 (within-person) predictors

For detailed examples of how we coded Level 1 singlehood variables, please refer to the supplemental materials.

Singlehood status was assessed at each visit, indicating whether someone was romantically partnered (coded as 0) or not (coded as 1). A visit was coded as ‘1’ (single) if participants reported any of the following during the past 6 months: 1) no romantic partnerships or 2) romantic partnerships that had ended before the visit, indicating they were currently single. Therefore, a ‘single’ status in this study was operationalized as the absence of ongoing and current romantic partnerships at the time of assessment.

Transition into singlehood was coded as ‘1’ for visits at which participants first became single (and remained single at the time of assessment) after being romantically partnered. This coding also captures transitions that occurred entirely within the interval of a given visit (e.g., if a participant entered and then ended a relationship during the 6-month interval of that visit). Conversely, transition out of singlehood was coded as ‘1’ for visits at which participants first entered a new romantic partnership (that was ongoing at the time of assessment) after being single. For both types of transitions, the timing of relationship formation or dissolution was considered. Specifically, the breakup or formation of a new partnership must have occurred within the interval between adjacent visits (6 months [180 days] for visits 1–6 and 12 months [365 days] for visit 7). Taken together, this coding reflects new transitions to the state (single or partnered) that participants were in at the time of assessment.

Singlehood length quantified the elapsed time youth reported being continuously single (no romantic partnership in the past 6 months) at each consecutive visit. For example, a participant who was single during 2 consecutive visits (Visits 2 and 3), had an interim partnership (Visits 4), and then returned to singlehood for 3 consecutive visits (Visits 5, 6, 7) would have ascending values of 1 and 2 for the first singlehood periods (Visits 2 and 3). Because of the interruption in singlehood at Visit 4 (coded as 0), the subsequent periods of singlehood at Visits 5, 6 and 7 would be coded as 1, 2, and 3, respectively. Interruptions in singlehood also included those within the same visit—that is, if a romantic partnership was reported within the last 6 months of that visit, regardless of current single status at the assessment. Additionally, in rare cases where participants missed intermediate visits, their singlehood length would restart at the next available visit. For example, if a participant was single at Visits 3 and 5 but missed Visit 4, the count would reset, with Visit 5 coded as ‘1’. Functionally, this coding approach is similar to a linear effect of time, assessing whether mental health changed with acclimation to singlehood.

Lastly, the effect of time (Occasion) was included both as a random slope and as a fixed main effect to index change over time in depression, anxiety, and stress across the study period. This variable was coded as 0–6, where 0 corresponded to the baseline visit and 6 corresponded to the last visit.

Level 2 (between-person) predictors

Level 2 predictors included proportion of singlehood periods (i.e., person mean of singlehood status), which represent a percentage of the total singlehood periods over the entire study (range=14.3%–100%), FOBS, and social support. All level 2 independent variables were treated as fixed and were included as person means (e.g., FOBS was aggregated across all available visits), centered by the grand mean of the sample.

Multilevel models

Model fitting was iterative. In the first set of models, to test unique effects of each singlehood dimension, all singlehood variables were included without the moderators (main effect model). All main effect models were run separately by outcome (depression, anxiety, stress) for a total of 3 models. Note that for better interpretations of the moderator’s main effects, we also ran the same main effect models with the inclusion of FOBS and social support (see supplemental materials). Then, interaction effects were included between the moderators and singlehood dimensions (interaction model) to examine their roles in the singlehood-mental health association. All interaction models were run separately by outcome and by moderator, resulting in a total of 6 additional models (3 outcomes × 2 moderators). Regarding statistical power, we conducted post hoc Monte Carlo simulations (see supplementals for more details). Overall, the power to detect effects varied greatly, but the power is consistently high for all significant effects, ranging from 0.88 to 1.00.

All main effect and interaction models controlled for participant age at baseline as well as both Level 1 and 2 gender identity. For gender identity, identifying as cisgender was coded as 1 and identifying as a gender minority (e.g., non-binary, transgender) was coded as 0 for the Level 1 variable. For the Level 2 variable, participants would be coded as gender minority (1) if they endorsed a non-cisgender identity at least once throughout their study participation, otherwise they were coded as cisgender (0). Additionally, in each interaction model, the main effect of the other moderator was included as an additional control.

Results

Table 2 shows descriptive statistics for continuous variables; all were normally distributed, and missingness ranged from 0% (e.g., social support) to 7.5% (Fear of Being Single². Bivariate correlations indicated moderate covariation between continuous constructs, with little evidence of extreme collinearity (Table 3). Thus, we proceeded with multilevel analyses.

Table 2.

Descriptive statistics for continuous variables.

	N	Min	Max	M	SD	Skew	Kurtosis
Anxiety	205	8.00	40.00	20.61	7.57	0.17	−0.43
Depression	205	8.00	40.00	19.79	7.60	0.22	−0.66
Stress	205	1.00	16.00	8.62	2.97	−0.42	−0.13
Social Support	205	1.00	7.00	5.05	1.21	−0.72	0.58
Fear of Being Single	188	1.00	5.00	2.65	1.24	0.18	−1.27
Proportion of Singlehood Periods	205	14.3	100.0	52.24	27.61	0.31	−1.06

Note. Because the FOBS measure was not administered at baseline and was only given when participants endorsed being single, descriptives for this variable reflect the person mean across all available occasions.

Table 3.

Bivariate correlations between continuous variables.

	1	2	3	4	5
1. Anxiety	--
2. Depression	.83**	--
3. Stress	.70**	.78**	--
4. Social Support	−.16*	−.33**	−.34**	--
5. Fear of Being Single	.52**	.44**	.38**	−.05	--
6. Proportion of Singlehood Periods	.20**	.23**	.24**	−.18*	.31**

Note. All variables were person means.

^**p<.01

Main Effect Models

In all main effect models (Table 4), the residual variance of the intercept (Est Range=3.31–25.01, ps<.001) and random variance of the slope of occasion were significant (Est Range=.07-.64, ps<.05), indicating that participants differed in their baseline values and trajectories of depression, anxiety, and stress. The significant effect of occasion suggested that stress (p=.04) decreased over time for SGM-AFAB youth in this sample.

Table 4.

Multilevel models for main effects

	Stress				Depression				Anxiety
	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p
Intercept	8.07	8.07	18.19	<.001	16.62	1.14	14.55	<.001	18.47	1.11	16.65	<.001
Age	−0.10	−0.10	−2.07	0.04	−0.24	0.12	−2.06	0.04	−0.28	0.13	−2.14	0.03
Gender Identity – Level 1	−0.47	−0.47	−1.56	0.12	−0.15	0.73	−0.20	0.84	−0.77	0.77	−0.99	0.32
Gender Identity – Level 2	0.31	0.31	0.84	0.40	2.36	0.91	2.60	0.01	2.12	0.96	2.20	0.03
Occasion	−0.09	−0.09	−2.02	0.04	−0.15	0.11	−1.37	0.17	−0.06	0.10	−0.55	0.58
Singlehood Status	0.46	0.46	1.38	0.17	1.57	0.73	2.14	0.03	1.16	0.69	1.69	0.09
Singlehood Length	−0.19	−0.19	−2.34	0.02	−0.51	0.19	−2.73	0.01	−0.28	0.19	−1.50	0.14
Transition into Singlehood	−0.16	−0.16	−0.73	0.47	−0.86	0.51	−1.69	0.09	−0.77	0.50	−1.54	0.12
Transition out of Singlehood	−0.01	−0.01	−0.04	0.97	−0.65	0.56	−1.16	0.25	−0.09	0.59	−0.15	0.88
Proportion of Singlehood Periods	0.02	0.02	3.20	0.001	0.04	0.02	2.78	0.01	0.04	0.02	2.39	0.02

Note. Significant effects are bolded.

For Aim 1, controlling for all other singlehood dimensions, youth who reported proportionally more periods of being single across the entire study on average reported more stress (p=.001), depression (p=.01), and anxiety (p=.02). At the within-person level, youth reported more depression at occasions when they were single (p=.03). In contrast, depression (p=.01) and stress (p=.02) decreased as singlehood length increased. There were no significant main effects of transitions into or out of singlehood.

Moderation Models

Interactions with FOBS and social support (Aim 2) are shown in Tables 5 and 6, respectively. Social support did not moderate any effects of singlehood dimensions. We also found no interaction between most of the singlehood dimensions and FOBS, except those between FOBS and singlehood length on stress (p=.01), depression (p<.001), and anxiety (p<.001). Simple slopes suggested that longer consecutive singlehood period was associated with decreased stress (Figure 1), depression (Figure 2), and anxiety (Figure 3) among youth with low FOBS (Stress Est=−.50, p=.001; Depression Est=−1.40, p<.001; Anxiety Est=−1.57, p<.001) and average FOBS (Stress Est=−.29, p=.001; Depression Est=−.75, p<.001; Anxiety Est=−.66, p<.001), but not for youth with high FOBS (Stress Est=−.08, p=.37; Depression Est=−.10, p=.61; Anxiety Est=.25, p=.21).

Table 5.

Multilevel models for FOBS interaction models

	Stress				Depression				Anxiety
	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p
Intercept	8.29	0.43	19.35	<.001	16.23	1.14	14.22	<.001	18.05	1.15	15.70	<.001
Age	−0.07	0.05	−1.37	0.17	−0.18	0.11	−1.63	0.10	−0.14	0.12	−1.16	0.25
Gender Identity – Level 1	−0.33	0.32	−1.00	0.32	0.12	0.79	0.15	0.88	−0.36	0.80	−0.46	0.65
Gender Identity – Level 2	0.10	0.38	0.25	0.80	1.89	0.98	1.93	0.05	2.08	0.99	2.09	0.04
Occasion	−0.08	0.05	−1.79	0.07	−0.14	0.11	−1.25	0.21	−0.04	0.10	−0.43	0.67
Singlehood Status	0.40	0.35	1.13	0.26	2.26	0.80	2.83	0.01	1.85	0.75	2.46	0.01
Singlehood Length	−0.29	0.09	−3.32	0.001	−0.75	0.19	−4.01	<.001	−0.66	0.18	−3.63	<.001
Transition into Singlehood	−0.31	0.23	−1.37	0.17	−1.05	0.51	−2.04	0.04	−1.16	0.49	−2.35	0.02
Transition out of Singlehood	−0.29	0.33	−0.89	0.38	−0.45	0.64	−0.70	0.48	−0.12	0.69	−0.17	0.87
Proportion of Singlehood Periods	0.01	0.01	1.63	0.10	0.01	0.02	0.90	0.37	0.01	0.02	0.74	0.46
Social Support	−0.76	0.15	−5.13	<.001	−1.86	0.52	−3.55	<.001	−0.85	0.49	−1.72	0.09
FOBS	0.37	0.20	1.82	0.07	0.76	0.51	1.49	0.14	1.42	0.58	2.44	0.02
FOBS*Singlehood Status	0.03	0.28	0.12	0.91	0.25	0.63	0.40	0.69	−0.40	0.62	−0.64	0.52
FOBS *Singlehood Length	0.17	0.07	2.55	0.01	0.52	0.13	4.09	<.001	0.74	0.14	5.30	<.001
FOBS*Transition into Singlehood	−0.06	0.18	−0.32	0.75	0.26	0.40	0.64	0.52	0.19	0.40	0.47	0.64
FOBS*Transition out of Singlehood	−0.06	0.25	−0.26	0.80	0.31	0.53	0.58	0.56	0.58	0.52	1.12	0.26
FOBS*Proportion of Singlehood Periods	−0.01	0.01	−0.97	0.33	−0.01	0.01	−0.82	0.41	−0.02	0.01	−1.70	0.09

Note. FOBS = fear of being single. Significant effects are bolded.

Table 6.

Multilevel models for social support interaction models

	Stress				Depression				Anxiety
	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p	Est	S.E.	Est/S.E.	p
Intercept	8.33	0.44	18.78	<.001	16.63	1.15	14.51	<.001	18.31	1.12	16.34	<.001
Age	−0.07	0.05	−1.44	0.15	−0.18	0.11	−1.68	0.09	−0.16	0.12	−1.27	0.21
Gender Identity – Level 1	−0.32	0.33	−0.98	0.33	0.08	0.79	0.11	0.92	−0.45	0.81	−0.55	0.58
Gender Identity – Level 2	0.05	0.38	0.12	0.91	1.69	0.97	1.74	0.08	1.89	0.99	1.92	0.06
Occasion	−0.09	0.05	−2.01	0.045	−0.17	0.11	−1.54	0.13	−0.07	0.11	−0.64	0.52
Singlehood Status	0.22	0.33	0.65	0.52	1.55	0.77	2.01	0.04	1.07	0.69	1.55	0.12
Singlehood Length	−0.19	0.08	−2.46	0.01	−0.48	0.19	−2.58	0.01	−0.28	0.19	−1.52	0.13
Transition into Singlehood	−0.21	0.22	−0.94	0.35	−0.80	0.53	−1.52	0.13	−0.73	0.51	−1.43	0.15
Transition out of Singlehood	−0.30	0.31	−0.98	0.33	−0.64	0.61	−1.05	0.29	−0.30	0.61	−0.49	0.63
Proportion of Singlehood Periods	0.01	0.01	1.62	0.11	0.01	0.01	0.85	0.39	0.01	0.02	0.66	0.51
Social Support	−0.77	0.21	−3.64	<.001	−1.66	0.73	−2.28	0.02	−0.74	0.61	−1.21	0.23
FOBS	0.52	0.12	4.15	<.001	1.51	0.31	4.94	<.001	1.96	0.35	5.64	<.001
SS*Singlehood Status	0.04	0.31	0.14	0.89	0.01	0.77	0.02	0.99	−0.17	0.65	−0.26	0.80
SS*Singlehood Length	−0.03	0.09	−0.38	0.70	−0.16	0.19	−0.83	0.41	−0.02	0.22	−0.10	0.92
SS*Transition into Singlehood	−0.01	0.24	−0.05	0.96	−0.22	0.62	−0.36	0.72	0.20	0.56	0.35	0.73
SS*Transition out of Singlehood	−0.12	0.25	−0.48	0.63	−0.57	0.61	−0.93	0.36	−0.45	0.63	−0.71	0.48
SS*Proportion of Singlehood Periods	−0.01	0.01	−1.02	0.31	−0.02	0.02	−1.34	0.18	−0.01	0.02	−0.33	0.74

Note. FOBS = fear of being single. SS = social support. Significant effects are bolded.

Fig 1.

Interaction between Level 1 singlehood length and Level 2 fear of being single on stress

Fig 2.

Interaction between Level 1 singlehood length and Level 2 fear of being single on depression

Fig 3.

Interaction between Level 1 singlehood length and Level 2 fear of being single on anxiety

Regarding main effects, before adding interaction terms (see supplemental materials), higher social support was related to lower levels of stress and depression (ps<.001), whereas higher FOBS was related to higher levels of all mental health outcomes (ps<.001). Notably, the effects of proportion of singlehood periods disappeared (even in models without interactions; see supplemental materials). Meanwhile, the effects of current singlehood status and length remained and even became significant for anxiety in the FOBS interaction model. In the same model, transition into singlehood was associated with lower depression (p=.04) and anxiety (p=.02), suggesting that newly single individuals may experience a sense of relief from exiting a presumably unsatisfying partnership. However, given that the inclusion of interaction terms drastically alters the interpretation of main effects³, we urge caution in interpreting any main effects in the interaction models.

Discussion

Singlehood and Its Dimensions

Our findings revealed significant between- and within-person associations among various dimensions of singlehood and mental health problems in SGM-AFAB youth. However, the nature of these associations is not uniform across all dimensions.

First, proportionally more periods of being single were associated with higher levels of stress, depression, and anxiety (though these associations disappeared after adding the moderators). Additionally, participants reported increased depression at times they were single, suggesting the transitory impact of singlehood status. Put simply, not only are the total experiences of being single linked to poor mental health across individuals, but being single also corresponds with fluctuations in mental health problems over time within the same individuals. This may not be surprising given the key role romantic partners play in reducing loneliness and helping individuals manage their mental health. (Muscari & Fleming, 2019). Our study confirms previous findings regarding the link between singlehood status and mental health challenges, while also adding depth to the existing literature by teasing apart the unique effects of overall singlehood exposure and current status.

Interestingly, a continuous period of remaining single was associated with decreased mental health problems. It is possible that one become more accustomed to singlehood as time progresses, possibly as the negative effects of singlehood dissipate. It is also possible that SGM youth adapt positively to single life over time. Those who continue to remain single may derive mental health benefits from increased positive appraisal and satisfaction with singlehood experiences (Beckmeyer & Jamison, 2023), as well as a decreased desire for romantic relationships over time (Kislev, 2021). This adaptation could also reflect a crucial identity development process, wherein remaining single for longer periods gives SGM youth more time and freedom to explore and solidify their identity and preferences.

The non-significant effects of transitions in this study (at least before the inclusion of several interaction terms) are in contrast with past research that typically focuses on major life changes such as marriage or entering a new relationship in adults (e.g., Bühler et al., 2023; Wade & Pevalin, 2004). It is possible that transitioning between non-marital relationship statuses is a normal part of youth development and thus does not impact mental health as significantly as it does in adults, especially when other important dimensions of singlehood (e.g., length) are considered.

Fear of Being Single and Social Support

Only the interactions between singlehood length and FOBS were significant. Specifically, an extended period of continuous singlehood was associated with decreased mental health problems, but only for those with low to average FOBS. Although past research suggests that people generally grow more content with being single over time (Park et al., 2022), SGM-AFAB youth with a strong fear of being single may dwell on the negative aspects of singlehood (e.g., romantic loneliness; Adamczyk, 2016) making it harder for them to adapt to single life. Conversely, those with relatively low FOBS may be more likely to adjust to and enjoy the benefits of singlehood (e.g., personal freedom), thereby enhancing their mental well-being over time.

Although no other interactions were found, FOBS and social support each showed a unique and independent association with mental health problems. First, in line with prior research involving non-SGM adults (Girme, Park, et al., 2022), FOBS was associated with detrimental mental health outcomes. Although the findings may not be concerning for some given the young age of this sample (ample time for relationship formation), it is crucial to recognize that single SGM youth experience barriers to relationship formation on top of common SGM-related stressors (DeHaan et al., 2013). Given the societal pressures placed on queer individuals to conform to heteronormative relationship norms, many single SGM-AFAB youth may fear further marginalization beyond what they already experience. The prospect of being romantically alone could also signal, both to themselves and others, a lack of full acceptance for their authentic self. The robust link between FOBS and mental health challenges observed in our sample emphasizes the need for further research on FOBS within marginalized communities. While our study focused on SGM youth, the implications of FOBS as a risk factor for mental health issues may extend beyond this group. The absence of non-SGM youth in our study raises questions about whether FOBS impacts mental health similarly or differently across groups. Including broader youth populations in future research could provide valuable insights into the effects of FOBS on mental health.

The impact of social support on mental health—as observed in this study— is not new (e.g., McDonald, 2018), especially for SGM youth who tend to experience loneliness, rejection, and isolation (Gattamorta et al., 2022; McDanal et al., 2021). In this study, higher levels of perceived social support were uniquely linked to lower stress and depression, even when controlling for all dimensions of singlehood and FOBS. These findings highlight the importance of support from social networks (e.g., family, friends, the SGM community) for SGM youth’s mental health.

Limitations and Future Directions

The current study measured FOBS only when participants reported no partnerships. However, FOBS can be experienced by both singles and non-singles (Spielmann et al., 2013). To better understand the effect of FOBS on mental health, future longitudinal studies should measure FOBS at all time points regardless of one’s partnership status.

The data were drawn from a larger study and all analyses therefore were completely exploratory. Although a longitudinal design may allow for conclusions regarding temporal relationships, ‘temporal’ should not be interpreted as ‘causal’. For that, confirmatory research better designed for causational hypothesis testing is needed. Further, despite our careful considerations, the method we used to code multiple dimensions of singlehood remained somewhat arbitrary. Additionally, we acknowledge potential concerns regarding statistical power given our modest sample size. For example, we cannot reject the possibility that some of the null interaction findings are simply a result of inadequate power to detect cross-level interaction effects. While our findings are largely consistent with prior research (Ayala & Coleman, 2000; Hostetler, 2009; Kamp Dush & Amato, 2005; Park et al., 2022), more robust, well-powered replication studies are essential to gain full confidence in the findings.

Finally, similar to most studies with historically underrepresented people, the findings from our restricted sample of SGM-AFAB youth may not generalize to different populations such as non-SGM and SGM youth assigned male at birth. We also did not examine the influences of other demographics/identities (e.g., race, disability) that could intersect and shape singlehood experiences.

Conclusions

Singlehood is an inherently complex construct consisting of several dimensions. Its link to mental health is anything but simple. One of the most commonly shared goals in relationship science is to ‘prevent’ singlehood by focusing on the acquirement and retention of relationships. But not all singles are victims waiting to be rescued. According to Queer Theory, singlehood and queer histories “have overlapping experiences of marginalization and expectations of living in a heteronormative society” (Lavender-Stott, 2023, p. 4). In this sample of SGM-AFAB youth, certain dimensions of singlehood (e.g., status) are associated with greater mental health problems, whereas others (e.g., length) show the opposite patterns. Interaction findings suggest that youth with low (but not high) FOBS appear to derive benefits from their prolonged singlehood duration, experiencing a reduction in mental health problems over time. Understanding the moderating role of FOBS is essential for addressing the specific fears and anxieties that can intensify the difficulties of singlehood. Moreover, given the significance of social support as a protective factor for mental health in SGM youth, educators, counselors, and stakeholders should equip them with skills to build and strengthen their social connections. Overall, we advocate for an inclusive research perspective that acknowledges the complexities of singlehood and promotes a supportive environment for SGM individuals, considering the nuances of their experiences and the contexts that shape them.