Introduction

Contemporary literature has firmly established a consistent pattern wherein lesbian women, gay men, and bisexual (LGB) individuals tend to exhibit worse mental health outcomes when compared to their heterosexual counterparts (Hottes et al., 2016; King et al., 2008; Shenkman & Shmotkin, 2010). This discrepancy is commonly attributed to an enduring strain of chronic minority stress, as many LGB individuals experience prevalent stigma and discrimination due to their sexual minority status (Frost & Meyer, 2023; Meyer, 2003). However, this comparative approach mainly focuses on differences based on sexual identity. As such, it has faced criticism for neglecting other crucial facets of identity that significantly shape the experiences of LGB individuals and families (Fish & Russell, 2018). This includes aspects such as romantic relationship status and parenthood status, which have both become increasingly prominent and relevant to LGB identity due to advancements in legislation and fertility technology (Bos & Gartrell, 2020; Carone et al., 2021). Furthermore, when viewed through the lens of social normative theoretical framework, the establishment of a lasting relationship followed by parenthood emerges as a prevailing societal expectation for both heterosexual and sexual minority individuals (Tate, 2022, 2023). Hence, and in order to explore LGB mental health in a more comprehensive fashion, the present study examines the associations between couplehood (being in a romantic relationship), parenthood, and mental health indicators (i.e., depressive symptomatology and life satisfaction) among Israeli LGB individuals. Additionally, we aimed to investigate the potential indirect effect of couplehood/parenthood on mental health through internalized homonegativity, referring to the negative sentiments towards one’s own sexual orientation arising from the internalization of antigay stigma and prejudice (Hatzenbuehler, 2009; Meyer, 2003). This exploration provides valuable insight into the role of internalized homonegativity in mental health differences between LGB who are coupled or parents and those who are not. In doing so, it also contributes to the existing literature on the strengths and resilience of queer families (e.g., Power et al., 2010).

The association of couplehood and parenthood with the mental health of LGB individuals

Heterosexual couples, whether married or cohabitating, have consistently demonstrated higher levels of mental health and psychological well-being compared to single adults (Gómez-López et al., 2019; Kamp Dush & Amato, 2005). Similar patterns have been observed among LGB couples (Riggle et al., 2010; Shenkman et al., 2023c). It has been posited that close relationships with significant others can promote social integration and provide social and economic support, thereby acting as a buffer against less favorable life events and fostering mental health (Shenkman et al., 2022a; Umberson et al., 1996).

The transition to parenthood among heterosexual individuals is commonly associated with decreased levels of life satisfaction and positive emotions, as well as increased levels of depressive symptoms and meaning in life indicators, such as purpose in life and personal growth (Nomaguchi & Milkie, 2020). This phenomenon, often referred to as the “parenthood paradox,” is typically explained by an increase in the perceived meaning in life upon becoming parents, juxtaposed with challenges in daily functioning, such as marital conflicts related to division of labor and sleep deprivation (Nomaguchi & Milkie, 2003, 2020). In the case of gay fathers, however, commonly there is no evidence for the “parenthood paradox”. Several studies have repeatedly found that Israeli gay fathers reported higher levels of life satisfaction and meaning in life compared to both childfree gay men (Shenkman & Shmotkin, 2014; Shenkman et al., 2023c) and heterosexual fathers (Erez & Shenkman, 2016; Shenkman et al., 2018, 2020). This pattern of findings is often accredited to the socio-cultural climate in Israel, which values parenthood and family life, thus promoting a strong sense of achievement and social acceptance among LGB individuals who have overcome legal and bureaucratic obstacles to become parents (Shenkman & Shmotkin, 2020; Shenkman et al., 2022a, b). Similar explanations have been proposed in studies comparing the mental health of lesbian mothers to heterosexual mothers (Shenkman, 2018; Shenkman et al., 2023a, b), while specific comparisons in mental health between lesbian mothers and lesbian childfree women remains scarce (Assink et al., 2022).

The examination of the associations between couplehood status, parenthood status, and mental health among LGB individuals is of the utmost importance, since it relates to a population that still faces discrimination in legislation recognizing same-sex marriage and parenthood (Costa & Shekman, 2020; Salvati et al., 2020). Furthermore, there is a lack of research exploring the role of internalized homonegativity in contributing to mental health disparities between LGB couples or parents and LGB individuals who are not in romantic relationships or who are not parents. Studying these issues is especially relevant in the Israeli socio-cultural context.

The Israeli socio-cultural context

The investigation of couplehood and parenthood among LGB individuals within the context of Israel holds significant importance, given the country’s familial and pronatalist societal characteristics (Birenbaum-Carmeli & Dirnfeld, 2008). This is evident with Israel's high fertility rate (3.1 children per family), which surpasses other countries in the OECD (OECD, 2019), and with the comprehensive coverage of various reproductive technologies provided for by the country’s health services (Birenbaum-Carmeli & Dirnfeld, 2008). Therefore, parenthood in Israel seems to serve as a primary avenue for social acceptance and inclusion, regardless of sexual orientation (Shenkman et al., 2022b).

Nevertheless, Israeli LGB individuals face limited access to same-sex marriage, same-sex adoption, and until recently, surrogacy services (Costa & Shenkman, 2020; Shenkman, 2022). Such systematic discriminatory practices can negatively impact the well-being of sexual minorities (Hatzenbuehler et al., 2010). Furthermore, sexual minorities may internalize such discriminatory practices and stigmas, potentially lessening their own estimations regarding succeeding to become parents or establishing stable romantic relationships (Shenkman, 2012).

Moreover, the Israeli socio-cultural context exhibits a prevailing patriarchal nature and is influenced by masculine norms, particularly in the context of recurrent war conditions and mandatory miliary service (Sion & Ben-Ari, 2009). These norms, along with orthodox Jewish religion, tend to condemn sexual minority individuals and contribute to a hostile environment for LGB individuals (Shenkman & Shmotkin, 2013; Shenkman et al., 2021) thereby facilitating the development of internalized homonegativity. While taking these socio-cultural characteristics into consideration, it would be highly beneficial to better understand the indirect effect of couplehood/parenthood on mental health through internalized homonegativity.

Internalized homonegativity and mental health

Internalized homonegativity, also known as internalized homophobia or internalized stigma, refers to the inward adoption of society’s homophobic attitudes by LGB individuals (Hatzenbuehler, 2009; Meyer, 2003; Meyer & Dean, 1998). It encompasses negative global attitudes towards homosexuality, discomfort with disclosing sexual orientation, disconnection from other LGB individuals, and discomfort with same-sex sexual activity. It is also characterized by an inner conflict between same-sex attraction and an obligation towards heterosexual norms (Herek, 2004).

Research on internalized homonegativity and mental health has commonly adopted a minority stress perspective (Frost & Meyer, 2023; Meyer, 1995, 2003). This offers a valuable framework to comprehend how societal norms and attitudes impact the well-being of sexual minorities. Minority stress theory suggests that sexual minority individuals experience stigma due to their minority status in predominantly heterosexual societies. This results in persistently high levels of stress, leading to increased mental health risks. The theory distinguishes between two types of stress processes: distal and proximal. Distal stress processes encompass external experiences of rejection, prejudice, and discrimination. Through cognitive evaluations and internalization, these experiences acquire psychological significance and become inner-proximal stressors. Consequently, sexual minority individuals facing prejudice may internalize these messages, perceiving them as negative aspects of their self-image (Meyer, 1995). This internalized homonegativity, is associated with feelings of low self-esteem, shame, difficulties in relationships, and guilt (Herek et al., 1998; Meyer, 2003; Mohr & Fassinger, 2006).

A vast and globally diverse compilation of research has illustrated that internalized homonegativity is associated with mental health symptoms such as depression and anxiety (e.g., Newcomb & Mustanski, 2010), alongside lower levels of general well-being and life satisfaction (Frost & Meyer, 2009; Gómez et al., 2022; Plant et al., 2022; Szymanski et al., 2008; Wen & Zheng, 2019). These results highlight the significance of internalized homonegativity as a proximal stressor that may negatively impact the mental health and overall well-being of sexual minorities.

Internalized homonegativity, couplehood, and parenthood among sexual minority individuals

Internalized stigma experienced by LGB individuals suggests a perceived stereotypical inability to legitimately engage in intimate and enduring partnerships which in turn potentially impedes their romantic prospects (Meyer & Dean, 1998). Indeed, internalized homonegativity as a minority stressor has been frequently suggested to be linked with difficulties in the romantic relationships (e.g., Brown & Trevethan, 2010; Frost & Meyer, 2009; Thies et al., 2016). For LGB individuals, being involved in a romantic relationship can symbolize a partial conquest over intrinsic aspects of internalized homonegativity such as feelings of shame and self-devaluation, (Mohr & Fassinger, 2006). Consistent with this notion, studies have shown that gay men in intimate relationships report lower levels of internalized homonegativity (Liang & Huang, 2022; Meyer & Dean, 1998).

The association between internalized homonegativity and parenthood status among LGB individuals has received limited attention in research. However, one study found lower levels of internalized homonegativity among lesbian and bisexual mothers as opposed to childfree lesbian and bisexual women (Assink et al., 2022). This observation suggests that realizing parenthood aspirations among LGB individuals may be associated with a certain degree of self-acceptance and a willingness to challenge stereotypical beliefs that LGB individuals should not become, or may be unfit, parents (Tsfati & Ben-Ari, 2019). Such self-acceptance could be linked with lower internalized homonegativity, which, in turn may result in more favorable mental health outcomes. Yet, to the best of our knowledge, no previous studies have investigated the potential indirect effect of parenthood status on mental health through internalized homonegativity among LGB individuals. Additionally, there is a lack of research exploring the indirect effect of couplehood status on mental health through internalized homonegativity among LGB individuals. Examining these aspects could provide insights into the factors contributing to higher levels of mental health among sexual minority couples or parents compared to those who are not in a relationship or do not have children (Shenkman et al., 2023c). By doing so, it may contribute to the growing literature on the strengths and resilience of queer families (e.g., Power et al., 2010).

Research hypotheses

In light of the existing literature discussed above, we formulated the following hypotheses:

  1. 1.

    LGB individuals who are parents or in a relationship will report lower levels of depressive symptoms and higher levels of life satisfaction compared to LGB individuals who are not parents or who are not in a relationship, respectively.

  2. 2.

    LGB individuals who are parents or in a relationship will report lower levels of internalized homonegativity compared to LGB individuals who are not parents or who are not in a relationship, respectively.

  3. 3.

    Internalized homonegativity will be associated with higher levels of depressive symptoms and lower levels of life satisfaction.

  4. 4.

    Parenthood/couplehood will have an indirect effect on mental health, as indicated by depressive symptoms and life satisfaction, through internalized homonegativity. Specifically, lower levels of internalized homonegativity will account for lower levels of depressive symptomatology and higher levels of life satisfaction observed among LGB individuals who are parents or in a relationship, when compared to those who are not.

Method

Participant

Participants included 491 LGB cisgender Israelis aged 18–64 years (M = 33.51, SD = 9.09). Of these, 25.3% (n = 124) identified as lesbian, 56.6% (n = 278) as gay men, and 13.4% (n = 66) and 4.7% (n = 23) as bisexual women and men, respectively. Three hundred and twenty participants had a spouse (65.2%). All lesbian and gay participants with a spouse reported that their spouse’s gender was similar to their own. Among bisexual individuals with a spouse, 51.7% reported that their spouse’s gender was similar to theirs. One hundred and eighty participants had children (36.7%). Among LGB parents, the most common routes to parenthood included surrogacy, donor insemination, and adoption. Most of the participants were born in Israel (91.0%), live in a city (90.4%), are Jewish (96.4%), secular (71.9%), have university level education (86.2%), are employed (81.7%), white (88.4%), males (61.1%), with a mean score of 3.40 (SD = 0.92) on their self-rated economic status, indicating average income on a 5-point scale ranging from 1 (low economic status) to 5 (high economic status), had a mean score of 2.43 (SD = 1.33) on their self-rated importance of religion values, indicating barely important of religion values on a 6-point scale ranging from 1 (not at all) to 6 (extremely important), and had a mean score of 4.35 (SD = 0.74) on their self-reported health status, indicating good health on a 5-point scale ranging from 1 (bad health status) to 5 (very good health status). Table 1 presents the descriptive characteristics of the study groups and the coding categories for these variables.

Table 1 Sociodemographic characteristics of the study groups
Full size table

Measures

Sexual orientation

Sexual orientation was assessed through a self-report as follows: 1 (heterosexual woman), 2 (heterosexual man), 3 (lesbian woman), 4 (gay man), 5 (bisexual woman), 6 (bisexual man). This assessment is common in sexual minority research (e.g., Shenkman, 2021).

Relationship status

Relationship status was assessed through a self-report as follows: 0 (not in a relationship), 1 (in a relationship).

Parental status

Parental status was assessed through a self-report as follows: 0 (not a parent), 1(I am a parent).

Measure of Internalized Sexual Stigma for Lesbians and Gay Men (MISS-LG)

We used the Measure of Internalized Sexual Stigma for Lesbian and Gay Men (MISS-LG; Lingiardi et al., 2012) to assess homonegativity in our sample. Every item is rated using a 5-point Likert-type scale, ranging from 1 (totally disagree) to 5 (totally agree), e.g., “At work I pretend to be heterosexual” and “Sometimes I think that if I were heterosexual, I could be happier”. In the current study we administered a short version of this measure composed of five items that were found more suitable to a time-constrained survey. A preliminary study using the abbreviated version of the MISS indicated good internal consistency (Baiocco et al., 2010). The respondent’s score was the items’ mean rating, with higher scores referring to more internalized homonegativity. Cronbach’s alpha coefficients were 0.72 and 0.75 among participants who were parents and among participants who were child-free, respectively. Cronbach’s alpha coefficients were 0.69 and 0.79 among participants who were in a relationship and among singles participants.

Center for Epidemiologic Studies Depression Scale (CES-D)

CES-D was designed to assess self-reported symptoms associated with depression (Radloff, 1977). This measure consisted of 20 items describing major components of depressive symptomology. For each item, participants were asked to rate how often they felt or behaved in a certain way in the past week (e.g., “I had trouble keeping my mind on what I was doing” and “I felt that everything I did was an effort”). Ratings ranged from 1 (rarely or none of the time) to 4 (most or all of the time). The respondent’s score was the items' mean rating, with higher scores referring to more depressive symptoms. Cronbach’s alpha coefficients were 0.91 and 0.92 among participants who were parents and among participants who were child-free, respectively. Cronbach’s alpha coefficients were 0.91 and 0.92 among participants who were in a relationship and among participants who were single. This measure is commonly used for research and clinical purposes (Stansbury et al., 2006) and is widely used in Israel (e.g., Shenkman & Abramovitz, 2021; Shenkman et al., 2020).

Satisfaction With Life Scale (SWLS)

This measure was constructed to assess life satisfaction as the cognitive aspect of subjective well-being (Diener et al., 1985). The measure consists of five items referring to judgments of one's own life (e.g., “In most ways my life is close to my ideal”) and are rated on a scale of 1 (strongly disagree) to 7 (strongly agree). Cronbach’s alpha coefficients were 0.83 among participants who were parents as well as among participants who were child-free. Cronbach’s alpha coefficients were 0.82 and 0.83 among participants who were in a relationship and among participants who were single, respectively. This scale was shown to have highly positive psychometric properties (Shenkman et al., 2020).

Procedure

Questionnaires were administered in Israel between November 2022 and April 2023. Participants were recruited through online announcements posted on social media and through LGBTQ social events. Both sexual minority and heterosexual individuals were invited to participate voluntarily and anonymously in a survey researching parenthood in Israel. The recruitment announcements included a link to an online-web survey (hosted on Qualtrics), to which 1,538 accessed. The current study only involved the LGB individuals who fully answered survey items MISS-LG, CES-D, SWLS, and completed the sexual orientation, parental status and relationship status variables. Therefore, 1,047 individuals who did not completely answer the mentioned scales, or that did not identify as sexual minorities, were excluded from the analyses. The final sample consisted of 491 participants. All participants provided their consent to participate prior to completing the survey and were invited to contact the researchers when they finished should they have any questions. The study was reviewed and received approval by the Institutional Review Board at the first author’s institution.

Data analysis

Data analysis was conducted using SPSS 29. Power analysis conducted by G*Power software (version 3.1.9.2) showed that a sample of 402 participant would be adequate to explore the study's hypotheses. Data distribution was assessed for skewness and kurtosis values, with all study variables being in the acceptable ranges (i.e., skewness ± 2, kurtosis ± 7, West et al., 1995), indicating a normal distribution. Preliminary analyses aimed to identify potential covariates by examining differences between the parental status groups and between the relationship status groups in the demographic variables, using chi-square tests and t-tests. These analyses showed some significant differences in the demographic variables between parents and non-parents, and between those who were coupled and those who were not. Respectively, these include age (t[266] = -10.89, p < 0.001 and t[296] = -3.84, p < 0.001,), education level (χ2[1] = 12.40, p < 0.001 and χ2[1] = 14.41, p < 0.001), economic status (t[489] = -5.00, p < 0.001 and t[489] = -3.39, p < 0.001,), and religiosity (χ2[1] = 3.99, p = 0.046 and χ2[1] = 9.91, p = 0.002,). Furthermore, differences in sociodemographic variables as a function of parental status were found in gender (χ2[1] = 3.99, p = 0.010). Lastly, differences in sociodemographic variables as a function of relationship status were found in importance of religion values (t[489] = 2.26, p = 0.012). Due to such differences, the aforementioned variables were controlled for alongside self-rated health and employment status which were found to significantly correlate with the study variables.

To investigate the study hypotheses, the PROCESS macro (model 4; Hayes, 2018) was implemented, controlling for age, education level, economic status, religiosity, and importance of religion values, which significantly differed between groups, and self-rated health and employment status who were significantly correlated with the study variables. Gender was not controlled for as it was further explored as a potential moderator. To test the significance of the indirect effect, Hayes’s (2018) method was used to calculate 5,000 bootstrapped samples assessing the indirect effect of parental status or relationship status on depressive symptoms or life satisfaction respectively through internalized homonegativity. A significant indirect effect was established when the confidence interval (CI) of the indirect effect did not include zero. The AB cross-product test is extensively considered the best test for indirect effects, and it is commonly recommended over more traditional mediation approaches (Baron & Kenny, 1986; Hayes, 2018). Finally, moderated mediation analyses (models 7 and 14 in the PROCESS macro) were exploratory conducted to identify any potential differentiation in the indirect effect models as a function of gender.

Results

The results presented in Table 2 show that being a parent associated with a higher likelihood of having a romantic relationship. In addition, being a parent and having a relationship associated with higher life satisfaction, lower depressive symptoms, and lower internalized homonegativity. Moreover, levels of life satisfaction were linked with lower levels of depressive symptoms and lower levels of internalized homonegativity. Lastly, levels of depressive symptoms were linked with higher levels of internalized homonegativity.

Table 2 Pearson correlations between the main study variables
Full size table

In order to explore our hypotheses, mediation analyses (indirect effects) were conducted (controlling for age, education level, economic status, religiosity, importance of religion values, self-rated health, and employment status) to test whether LGB individuals who are parents/in a relationship had lower levels of depressive symptoms and higher levels of life satisfaction than LGB individuals who are childfree/not in a relationship and, if so, whether this disparity was accounted for by internalized homonegativity.

Along with our first hypothesis, that LGB individuals who are parents or in a relationship will report lower levels of depressive symptoms and higher levels of life satisfaction than those who are not parents or in a relationship, respectively, being parent or in a relationship was associated with lower levels of depressive symptoms (b = -0.17, p = 0.007 and b = -0.15, p = 0.002, respectively) and with higher levels of life satisfaction (b = 0.53, p = 0.000 and b = 0.55, p < 0.001, respectively). In addition, aligned with our second hypothesis, that LGB individuals who are parents or in a relationship will report less internalized homonegativity than those who are childfree or single, being a parent was associated with lower level of internalized homonegativity (b = -0.30, p = 0.003). Similarly, being in a relationship was associated with lower level of internalized homonegativity (b = -0.29, p = 0.000).

In line with our third hypothesis, internalized homonegativity associated with higher level of depressive symptoms while controlling for parenthood and while controlling for relationship status (b = 0.11, p = 0.000 and b = 0.11, p = 0.000, respectively). In addition, internalized homonegativity associated with lower levels of life satisfaction while controlling for parenthood and while controlling for relationship status (b = -0.38, p < 0.001 and b = -0.36, p < 0.001, respectively).

Finally, as seen in Figs. 1 and 2, and corresponding with our fourth hypothesis, the CIs of the indirect effects of parenthood on depressive symptoms or life satisfaction through internalized homonegativity did not contain zero (mediated b = -0.03, SE = 0.01, 95% CI [-0.12, -0.02] and mediated b = 0.11, SE = 0.04, 95% CI [0.04, 0.19], respectively). This presents a partial indirect effect of parental status on depressive symptoms and life satisfaction through internalized homonegativity.

Fig. 1
figure 1

The indirect effect of parental status on depressive symptoms (a) and life satisfaction (b) through internalized homonegativity. Note. N = 491. Reported values are unstandardized regression coefficients (bs) for pathways among parental status (0 = childfree, 1 = parent), internalized homonegativity, depressive symptoms and life satisfaction. Level of education, age, economic status, health, employment status, importance of religious values and religiosity served as covariates. The total effects of parental status on depressive symptoms or life satisfaction are reported in parenthesis. *p < .05 ** p < .01. *** p < .001

Full size image
Fig. 2
figure 2

The indirect effect of relationship status on depressive symptoms (a) and life satisfaction (b) through internalized homonegativity. Note. N = 491. Reported values are unstandardized regression coefficients (bs) for pathways among relationship status (0 = single, 1 = in a relationship), internalized homonegativity, depressive symptoms and life satisfaction. Level of education, age, economic status, health, employment status, importance of religious values and religiosity served as covariates. The total effects of relationship status on depressive symptoms or life satisfaction are reported in parenthesis. *p < .05 ** p < .01. *** p < .001

Full size image

Similar indirect effect was found for relationship status, as the CI of the indirect effect of relationship status on depressive symptoms and life satisfaction through internalized homonegativity did not contain zero (mediated b = -0.03, SE = 0.01, 95% CI [-0.06, -0.01], and mediated b = 0.10, SE = 0.03, 95% CI [0.04, 0.17], respectively). The results indicate a partial indirect effect of relationship status on depressive symptoms or life satisfaction through internalized homonegativity.

Notably, given the correlation between parental status and relationship status, we reanalyzed the data for our four hypotheses, controlling for relationship status in all analyses related to parental status. Additionally, we conducted a second set of analyses for our hypotheses, controlling for parental status in all analyses focusing on relationship status. The results revealed that when parceling out the effects linked to relationship status from those of parenthood status and vice versa, the pattern of results remained consistent to the results descried above. For example, the CIs of the indirect effects of parenthood did not contain zero, both when predicting depressive symptoms and life satisfaction, after controlling for relationship status, age, education level, economic status, religiosity, importance of religion values, self-rated health, and employment status (mediated b = -0.02, SE = 0.01, 95% CI [-0.05, -0.00] and mediated b = 0.73, SE = 0.04, 95% CI [0.01, 0.15], respectively). Similarly, the CI of the indirect effect of relationship status on depressive symptoms and life satisfaction through internalized homonegativity, after controlling for parental status, age, education level, economic status, religiosity, importance of religion values, self-rated health, and employment status, did not contain zero (mediated b = -0.03, SE = 0.01, 95% CI [-0.05, -0.01] and mediated b = 0.08, SE = 0.03, 95% CI [0.02, 0.13], respectively). All these additional analyses are available upon request from the first author.

To examine exploratory whether the mediation models (indirect effects) differed as a function of gender (i.e., men vs. women), moderated mediation analyses were conducted, focusing on the possible moderating effect of gender in the association between parental status and internalized homonegativity, and in the association between relationship status and internalized homonegativity (i.e., path A, model 7 in the PROCESS macro). The CIs of the indices of moderated mediation (Hayes, 2018) contained zero when predicting depressive symptoms or life satisfaction with parental status as a predictor and internalized homonegativity as a mediator (index = -0.02, SE = 0.02, 95% CI [-0.05, 0.12] and index = 0.05, SE = 0.06, 95% CI [-0.06, 0.17], respectively). Similarly, the CIs of the indices of moderated mediation also contained zero when predicting depressive symptoms or life satisfaction with relationship status as a predictor and internalized homonegativity as a mediator (index = 0.01, SE = 0.02, 95% CI [-0.03, 0.05] and index = -0.02, SE = 0.06, 95% CI [-0.15, 0.11], respectively). Thus, the indirect effects were unrelated to gender.

Additional moderated mediation analyses examined the possible moderation of gender in the association between internalized homonegativity and either depressive symptoms or life satisfaction (path B, model 14 in the PROCESS macro). The CIs of the indices of moderated mediation (Hayes, 2018) contained zero when predicting either depressive symptoms or life satisfaction with parental status as a predictor and internalized homonegativity as a mediator (index = -0.01, SE = 0.02, 95% CI [-0.05, 0.03] and, index = -0.00, SE = 0.04, 95% CI [-0.08, 0.07] respectively). Similarly, the CIs of the indices of moderated mediation contained zero when predicting either depressive symptoms or life satisfaction with relationship status as a predictor and internalized homonegativity as a mediator (index = -0.01, SE = 0.02, 95% CI [-0.04, 0.03], and index = 0.00, SE = 0.04, 95% CI [-0.07, 0.07], respectively). So too these indirect effects were unrelated to gender.

To explore alternative models, particularly the indirect effects of parental or relationship status on internalized homonegativity through mental health (depressive symptoms or life satisfaction), we conducted supplementary (indirect effects) analyses.

Initially, we examined the indirect effect of parental status on internalized homonegativity through depressive symptoms, controlling for age, economic status, employment status, importance of religious values, religiosity, education level, and relationship status. The results revealed that the confidence interval (CI) of the indirect effect contained zero (mediated b = -0.03, SE = 0.02, 95% CI [-0.07, 0.01]), indicating that there was no significant indirect effect. Subsequently, we examined the the indirect effect of parental status on internalized homonegativity through life satisfaction. The results revealed that the CI of the indirect effect did not include zero (mediated b = -0.06, SE = 0.03, 95% CI [-0.12, -0.01]), indicating an indirect effect of parental status on internalized homonegativity through life satisfaction.

Additionally, we examined the indirect effect of relationship status on internalized homonegativity through depressive symptoms, controlling for age, economic status, employment status, importance of religious values, religiosity, education level, and parental status. The results revealed that the CI of the indirect effect did not include zero (mediated b = -0.03, SE = 0.02, 95% CI [-0.06, -0.00]), indicating an indirect effect of relationship status on internalized homonegativity through depressive symptoms. Similarly, we examined the indirect effect of relationship status on internalized homonegativity through life satisfaction. The results revealed that the CI of the indirect effect did not include zero (mediated b = -0.08, SE = 0.03, 95% CI [-0.14, -0.04]), indicating an indirect effect of relationship status on internalized homonegativity through life satisfaction. These findings from alternative models yielded generally weaker results compared to the original model, particularly given the absence of an indirect effect of parental status on internalized homonegativity through depressive symptoms in the alternative model.

Discussion

Consistent with our first and second hypotheses, LGB individuals who were parents or in a relationship reported lower levels of depressive symptoms, higher levels of life satisfaction, and lower levels of internalized homonegativity compared to LGB individuals who were not parents or who were not in a relationship, respectively. Supporting our third hypothesis, internalized homonegativity was found to be positively associated with higher levels of depressive symptoms and negatively associated with life satisfaction. Furthermore, confirming our fourth hypothesis, parenthood/couplehood had an indirect effect on mental health through internalized homonegativity, such that the lower levels of depressive symptoms and higher levels of life satisfaction observed among LGB individuals who are parents or who are in a relationship, compared to those who are not, were partially explained by their lower levels of internalized homonegativity.

LGB individuals who were parents or in a relationship exhibited lower levels of depressive symptoms and higher levels of life satisfaction, in comparison to LGB individuals who were not parents or who were not in a relationship, respectively. These findings align with previous research indicating that being in a romantic partnership can provide both financial and emotional support, along with a sense of conformity to societal expectations of forming a couple, particularly in societies that prioritize family values (Kamp Dush & Amato, 2005; Shenkman & Shmotkin, 2014). Such factors may contribute to enhanced mental health (Shenkman et al.,, 2023c). Similarly, the positive mental health outcomes observed among LGB parents compared to LGB individuals without children may stem from a sense of achievement following the navigation of numerous bureaucratic, legal, and financial obstacles encountered on the path to LGB parenthood. This sense of accomplishment, especially within pronatalist sociocultural contexts as in Israel, may be associated with heightened levels of well-being (Shenkman et al., 2022b). Identifying the protective role of couplehood and parenthood in the realm of mental health for LGB individuals is in line with current calls to explore factors contributing to the well-being of sexual minority individuals, without relying solely on comparisons to heterosexual control groups (Fish & Russell, 2018). It also aligns with an expanding body of research that centers on the resilience and strengths observed in families led by same-gendered parents (Power et al., 2010; Shenkman et al., 2023a, b, c).

As predicted, LGB individuals involved in a romantic relationship exhibited lower levels of internalized homonegativity compared to those who were not in a relationship. Consistent with previous research linking romantic relationships to reduced internalized homonegativity (Frost & Meyer, 2009; Meyer & Dean, 1998), it could be argued that maintaining a romantic partnership among LGB individuals signifies a triumph over internalized homonegative stigmas suggesting that they are incapable of intimacy, healthy romantic connections, or deserving of such relationships (Liang & Huang, 2022). Our current results corroborate prior findings by also adding equivalent results from the context of Israel.

Our findings also revealed that LGB individuals who were parents had lower levels of internalized homonegativity in comparison to LGB individuals without children. In comparison with a recent study conducted in the United States where lesbian and bisexual mothers reported lower levels of internalized homonegativity compared to childfree lesbian and bisexual women (Assink et al., 2022), our study contributes new insights by generalizing these findings to gay and bisexual individuals as well, and by examining these associations in a non-Euro-American context. These findings support the notion that pursuing and achieving parenthood within the framework of LGB identity may require a certain level of self-acceptance and overcoming the prevailing stigma that LGB individuals should not be parents (Carone et al., 2021; Farr & Vázquez, 2020). Such processes seem to associate with lower levels of internalized homonegativity (Brown & Trevethan, 2010), and were rarely studied thus far in the context of LGB parenthood.

Consistent with our third hypothesis, a higher level of internalized homonegativity was found to be associated with increased depressive symptomatology and decreased life satisfaction. These findings follow a body of existing research that has consistently demonstrated the link between internalized homonegativity and adverse mental health outcomes, including higher levels of depression, anxiety, and lower levels of well-being (Frost & Meyer, 2009; Gómez et al., 2022; Newcomb & Mustanski, 2010; Plant et al., 2022). These results highlight the role of internalized homonegativity as a proximal stressor (Frost & Meyer, 2023) that may negatively affect the mental health of sexual minority individuals. For example, internalized homonegativity is frequently associated with reduced self-esteem and heightened experiences of shame, guilt, and self-hatred (Herek et al., 1998; Meyer, 2003) which is associated with adverse mental health outcomes. Notably, our results revealed a stronger association between internalized homonegativity and life satisfaction compared to the association between internalized homonegativity and depressive symptoms. This pattern aligns with previous research demonstrating similar pattern of correlations (e.g., Gómez et al., 2022). The relatively weaker effect observed between internalized homonegativity and depressive symptoms could potentially be attributed to the relatively young age of the LGB individuals in our sample. Previous meta-analytic review has indicated that studies with a younger mean age tend to exhibit weaker associations between internalized homonegativity and internalizing mental health problems, such as depression (Newcomb & Mustanski, 2010). Moreover, while depressive symptoms serve as a fundamental indicator of internalizing mental health problems, life satisfaction is a key component of more general subjective well-being (Keyes et al., 2002). It could be postulated that a broader, global assessment of overall well-being, such as life satisfaction, is more susceptible to the impact of internalized homonegativity than more specific indicators of adverse mental health, like depressive symptoms. However, further exploration of this rationale is warranted in future investigations.

Furthermore, the present findings highlight the indirect effect of couplehood/parenthood on mental health through internalized homonegativity. These findings illuminate possible underlying mechanisms that contribute to the disparities in depressive symptoms and life satisfaction observed between LGB individuals who are parents or who are in a relationship and those who are not. It can be speculated that maintaining a romantic partnership can represent a greater self-acceptance of sexual minorities’ sexual orientation and a partial conquest over ingrained facets of internalized homonegativity, which often insinuates that LGB individuals are incapable of experiencing genuine intimacy (Liang & Huang, 2022). This lower levels of internalized homonegativity in turn, may lead to better mental health outcomes (Gómez et al., 2022) indicated in the current study by lower levels of depressive symptoms and higher levels of life satisfaction.

Similarly, the present findings of the indirect effect of parenthood on mental health through internalized homonegativity, may suggest that becoming an LGB parent may be linked with greater self-acceptance and a readiness to confront stereotypical and homophobic notions that LGB individuals are unsuitable or incapable of parenthood (Tsfati & Ben-Ari, 2019). By challenging these prejudiced views, individuals may cultivate lower levels of internalized homonegativity, thereby fostering more positive mental health and well-being outcomes.

The observed indirect effects contribute to the existing research aimed at enhancing our understanding of mental health disparities within the sexual minority population (Assink et al., 2022; Shenkman & Shmotkin, 2014). Moreover, these findings can contribute to the minority stress theory (Frost & Meyer, 2023; Meyer, 1995, 2003, 2014), suggesting a more nuanced link between the proximal stressor of internalized homonegativity, mental health, couplehood, and parenthood among sexual minority individuals.

Strengths and limitations

A key strength of the present study lies in its novel examination of the indirect effect of couplehood/parenthood on mental health through internalized homonegativity among a sample of LGB individuals. It contributes innovative insights to the field of mental health among sexual minorities by considering the unique role of couplehood and parenthood status. However, several limitations should be acknowledged. Although we adhered to the guidelines of targeted sampling (Watters & Biernacki, 1989), our sample was not derived from a random or representative selection. Additionally, the fact that most participants were educated, employed, secular, and from urban areas, may restrict the ability to generalize the results to the entire LGB population. As the assessment of sexual orientation is not included in national surveys in Israel at present, obtaining a representative sample of LGB individuals is currently unfeasible. Moreover, the study was conducted in Israel, limiting the generalizability of the findings to other socio-cultural contexts. It is recommended to conduct similar studies in different countries to strengthen the external validity of the findings. Furthermore, our study relied on self-reported measures, which are subject to potential biases related to self-presentation. The relatively restricted alpha coefficients of the internalized homonegativity scale should also be considered. The correlational design of the study precludes making causal inferences; therefore, the bi-directionality of mediations/indirect effects should be taken into consideration (Maxwell et al., 2011). For instance, an alternative model could posit that parenthood or relationship status is associated with increased life satisfaction and decreased depressive symptoms, which, in turn, associates with reduced one's sense of internalized homonegativity. Exploring this alternative model (refer to exploratory analyses in the results section) yielded generally weaker results compared to the original model (particularly given the absence of an indirect effect of parental status on internalized homonegativity through depressive symptoms in the alternative model), thus supporting the main model. Nevertheless, while our main theoretical model and the positioning of the study variables in the model were guided by prior research focusing on elucidating disparities in mental health among sexual minority individuals based on relationship or parenthood status (e.g., Shenkman et al., 2023c), a longitudinal study is necessary for a more robust examination of causal pathways and additional alternative models. Mental health was assessed using both negative indicators (e.g., depressive symptomatology) and positive indicators (e.g., life satisfaction), in line with common practices (e.g., Shenkman & Shmotkin, 2020). However, it is important to explore additional indicators of mental health to provide a comprehensive understanding of this multifaceted construct. While our current findings did not indicate gender differences in the reported models, it is important for future studies to further investigate potential variations among LGBTQ + sub-groups using larger sample sizes. Despite numerous similarities, differences exist between lesbian couples/parents and gay couples/parents regarding their experiences of vulnerabilities and strengths in both couple relationships and parenthood (Cao et al., 2016). Similarly, bisexual individuals, and especially those with opposite-gender partners, may be less impacted by internalized homonegativity as they may pass as heterosexual individuals. While the current sample size of these sub-groups may be too small to examine this reasoning empirically, larger samples from each sub-group within the sexual minority population will allow for the exploration of the suggested model separately for lesbian women, gay men, bisexual individuals (with same, and opposite-gender partners), transgender individuals, and other queer populations. Variables such as social support and perceived stigma and discrimination, which could act as potential intervening variables in the relationship between the study variables, were not assessed in our study. Including these variables in future research may provide a more comprehensive understanding of the interplay between couplehood, parenthood, internalized homonegativity, and mental health. Likewise, it is strongly recommended to assess more nuanced variables pertaining to relationship status and parental status, such as the quality of couple relationships and parenting stress. Overall, the methodological limitations mentioned reflect common challenges encountered in investigations involving sexual minority populations (e.g., Krueger et al., 2020).

Conclusion and implications

The findings of this study indicate that both being in a relationship and being a parent are linked to better mental health in LGB individuals. Additionally, it was observed that the associations of these factors with mental health is partly accounted by internalized homonegativity. To advance this research area, future studies should investigate other relevant variables to this indirect effect, including satisfaction with romantic relationships and parenthood, perceived social and family support, and social norms such as familism and pronatalism. Furthermore, it is important to extend this model to explore the experiences of transgender individuals and other gender identities. To ensure a comprehensive understanding, larger sample sizes should be employed to include underrepresented groups, such as bisexual individuals, which were limited in our study due to small sample sizes. These results have significant implications for policy makers and legislators, highlighting the importance of recognizing the role of couplehood and parenthood in promoting positive mental health outcomes among sexual minority individuals who still face stigma, discrimination, and harassment (Frost & Meyer, 2023; Salvati et al., 2020). To address these issues, policy makers and legislators should work towards legalization of marriage and parenthood for sexual minority individuals and promoting inclusive policies that accommodate diverse family structures. This is particularly relevant in the Israeli socio-cultural context, where same-sex marriage remains illegal and paths to parenthood for sexual minorities are subject to social controversy. Mental health professionals and social agents should be familiar with these findings, as they underscore the association between internalized homonegativity and depressive symptoms, as well as an even stronger link between internalized homonegativity and impairment in general subjective well-being indices as life satisfaction. In addition to clinical interventions aimed at addressing the internalization of these stigmas, efforts to promote social change by policymakers and reduce the stigmatization experienced by sexual minorities—often internalized by individuals within this community—can significantly improve the mental health and well-being of sexual minority individuals.