Coming out Experiences and Disclosure gap in Three Age Cohorts of Portuguese Cisgender Sexual Minority Men

Studies emphasize that the time elapsed between self-awareness of one’s sexual orientation and its disclosure to others (disclosure gap) can be an indicator of psychosocial adjustment of sexual minorities. This study examined the coming out experiences and disclosure gap of three generations of sexual minority men: adults, 25–39 years; middle-aged, 40–59 years; and seniors, 60 or more years. A sample of 274 cisgender men was recruited, with ages ranging from 25 to 79 years. Data was collected between 2018 and 2019. Although most men had disclosed their sexual orientation to significant others, there was a wide variation on the coming out timing and experiences across the three age cohorts. Senior sexual minority men realized and disclosed their sexual orientation later than middle-aged men, while the latter realized and disclosed later than their younger counterparts. The disclosure gap followed a similar trend increasing with age. Regression analyses revealed that high levels of self-stigma explained a larger disclosure gap among the middle-aged, whereas low community connectedness explained a larger disclosure gap among seniors. No significant predictors emerged among adults. Overall coming out experiences worsened with age, with the older cohort reporting realizing and disclosing their sexual orientation later in life, taking longer to come out after identifying as sexual minorities, and feeling less accepted by others after coming out. This study highlighted some of the social factors that may improve sexual minority’s psychosocial well-being and possibly counterbalance the negative effects of stigma, namely, connection to the LGBT community.


Introduction
The coming out process has been widely studied among sexual minorities, including becoming aware of one's sexual orientation and disclosing it to significant others (Dunlap, 2016;Frost & Meyer, 2012;Meyer et al., 2021). Six main developmental milestones for a sexual minority identity have been identified, namely, becoming aware of one's same-sex attractions, questioning one's sexual orientation, same-sex sexual activity, same-sex romantic relationship, self-identification with a sexual minority identity, and disclosing one's sexual orientation or identity to others. Although the sequence and pacing of these milestones differs based on several aspects (e.g. gender, age cohort, sexual identity label), this process usually starts with becoming aware of one's same-sex attraction and end with disclosing one's sexual orientation to others (Bishop et al., 2020;Hall et al., 2021;Savin-Williams & Cohen, 2015).
This scholarly work suggests that one of the fundamental components of the coming out process -the disclosure of one's sexual orientation -is an important milestone in the development and integration of a sexual minority identity, regardless of the age at which it occurs (Dunlap, 2016). However, some studies have highlighted that the coming out process is complex and highly differentiated and intersubjective, that it may not necessarily happen in a single moment, and that the disclosure of sexual orientation and consequent response from others is one of the greatest challenges during the coming out process Grierson & Smith, 2005). In fact, the disclosure can occur only partially (e.g. disclosing to siblings but not to parents), only in some contexts (e.g. disclosing to friends but not to family members), and at several moments throughout the life cycle Grierson & Smith, 2005;Pereira et al., 2019). Nevertheless, the recipients of the first disclosure are usually close friends or other sexual minorities, whereas the disclosure to parents happens more often at later stages of the coming out process (e.g. Savin- Williams & Cohen, 2015).
Thus, a dimension that is becoming increasingly considered as essential for the psychosocial adjustment of sexual minorities is the time elapsed between becoming aware of same-sex attraction and/or sexual orientation and its disclosure to others (Guittar & Rayburn, 2015). This disclosure gap may span several months or years, and it may represent a strategy of self-protection and identity integration through the achievement of other important milestones and fostered by strong and supportive social networks and connectedness with the LGBT community, which in turn may lead to a better psychosocial adjustment (Corrigan & Matthews, 2003;Ribeiro-Gonçalves et al., 2019). In contrast, the disclosure gap may be prolonged by self-stigma and expectations of rejection by loved ones, which may cause psychological suffering and distress (Corrigan & Matthews, 2003;Costa et al., 2013;Guittar & Rayburn, 2015;Lyons & Pepping, 2017;Williams & Fredriksen-Goldsen, 2014). The coming out process may be strongly influenced by external factors, and the interaction of social, environmental, historical, and psychological factors determine wide variations in the disclosure gap for different generations of sexual minorities (Dunlap, 2016;Grierson & Smith, 2005;Pereira & Monteiro, 2016). Empirical studies and systematic reviews have shown that younger cohorts tend to achieve most of the sexual orientation milestones earlier than older cohorts, namely, same-sex sexual activity and romantic relationships, as well as disclosing their sexual orientation to others (Bishop et al., 2020;Savin-Williams & Cohen, 2015). However, there is contrasting evidence regarding whether there are differences in the timing of becoming aware of one's sexual orientation (Bishop et al., 2020;Hall et al., 2021). If younger cohorts do achieve all milestones earlier and have an accelerated coming out process than older cohorts, it suggests smaller disclosure gaps among younger generations of sexual minority men.
Generational and cohort studies allow a more complete and integrated perspective of the coming out process of sexual minorities against the backdrop of significant sociopolitical changes in the affirmation of LGBT rights (Dunlap, 2016). Further, these studies may also inform the preparation of appropriate and effective psychosocial and health interventions targeted at different age groups (Accornero, 2014;Dunlap, 2016;Pereira & Monteiro, 2016). Studies suggest that sexual minority men become aware of their sexual orientation and tend to disclose it to others earlier than sexual minority women across different age cohorts and that the association between disclosure and mental health is different for sexual minority men and women, suggesting somewhat different coming out experiences between genders (e.g. Dunlap, 2016;Pachankis et al., 2015). In addition, younger cohorts tend to disclose their sexual orientation earlier than their older counterparts, but also among men across different age cohorts, a smaller disclosure gap has been reported (e.g. Grov et al., 2006). Considering this evidence, in this study, we decided to examine the coming out experiences, with a particular focus on the most consistently reported first and last sexual orientation milestones (awareness and disclosure), only among cisgender sexual minority men. This study purports to examine the coming out process, experiences, timing of awareness of one's sexual orientation, timing of disclosure of sexual orientation, and disclosure gap of three generations of sexual minority cisgender men, divided into developmental and generational cohorts (25-39 years = adulthood, 40-59 years = middle age, 60 or more years = senior age; South, 2017; United Nation (UN), 2015).

Study Context
For close to five decades, Portuguese sexual minorities lived under radical oppression under a dictatorship regime (from 1932 to 1974), which makes the Portuguese socio-political context to some extent unique in comparison to other Western countries. The dictatorship regime greatly impacted and delayed the industrialization of Portugal, and most of the population was poor and uneducated, as they were mostly focused on fulfilling their basic needs such as having food and a home (Brandão, 2008). According to Portuguese official census, by 1970 more than 25% of the population was illiterate. Recent data further shows that close to 20% of the older population (60 + years; born before 1960) never entered the education system and about 50% completed only 4 years of formal education (Pordata, 2021). As such, senior sexual minorities from Portugal grew up struggling with poverty, access to education and employment, in addition to radical oppression of their same-sex behaviours. Even after the dictatorship ended, a decade of extreme political turmoil followed, further delaying the LGBT organized movement. Unlike in the USA, where the modern LGBT movement started in 1969 with the Stonewall riot, and the first Pride was organized the following year, in Portugal the first Pride march only took place in 2000 (Almeida, 2009).
Our second cohort's -middle-aged sexual minority men (40-59 years of age; born between 1960 and 1980) -coming of age was marked by a slow opening to a globalized world, namely, Portugal entering the European Economic Community (now European Union) in 1986. Consequently, it was only in the late 1980s that Portugal found economic growth and political stability. However, LGBT rights were completely absent from public or political discourses, and for 20 years after the decriminalization of homosexually in 1982, no other legal advances happened. In addition, the influence of the Catholic church extended beyond the end of the dictatorship regime, further oppressing sexual minority people (Santos, 2018). The LGBT movement was virtually inexistent during the 1970s and 1980s, and it was only in the 1990s that significant LGBT activism emerged, and the first LGBT organization -ILGA-Portugal -was created in 1995 to respond to the AIDS pandemic (Almeida, 2009;Brandão, 2008). Under these circumstances, this cohort of sexual minority men were mostly silent and silenced, with no formal LGBT community or resources available, and living their sexuality in secrecy.
In sharp contrast with the previous two generations, the last cohort of adult sexual minority men (25-39 years; born after 1981) grew up in a fast-changing Portugal, and similarly to the US Equality generation, this cohort participated in the LGBT movement that fought for LGBT rights and achieved the most important legal milestones during their coming of age: civil unions in 2001, non-discrimination laws in 2004, same-sex marriage in 2010, gender identity affirmation in 2011, and same-sex parenting in 2015 (Costa, 2021;Pereira & Monteiro, 2016). Noteworthy, progress regarding LGBT rights and protections was very fast paced in a country with such a recent history of LGBT activism.
These age cohorts differ from those proposed by the Generations Study in the USA (Bishop et al., 2020;Meyer et al., 2021): 18-25 years (equality cohort), 34-41 years (visibility cohort), and 52-59 years (Pride cohort) in two aspects: 1. We included the older generation (60 + years), the most invisible generation of sexual minorities that have only recently gained scientific attention (Fredriksen-Goldsen, 2018), which is absent from the Generations Study.

Sociocultural milestones of the Pride and Visibility
Generations in the USA differ from those in Portugal, whereas the Pride cohort were marked by the early days of the LGBT rights movement in the USA; this generation was still under severe oppression by an authoritative political regime in Portugal.
Therefore, Pride came alongside the HIV/AIDS pandemic in the mid-to late 1980s, effectively merging the Pride and the Visibility generations into one in Portugal. The younger cohort in this study is characterized by Equality, in which their coming of age was marked by growing social acceptance and human rights legislation, which makes this the only cohort directly comparable to that of the Generations Study.

Coming Out Experiences Across Different Age Cohorts
The way sexual minority individuals experience the coming out process varies widely depending upon their context and historical time (Dunlap, 2016). Coming out may reflect an integrated sexual minority identity and is generally associated with low levels of self-stigma and positive self-directed attitudes (Pereira & Leal, 2005;Rosario et al., 2006). However, some sexual minority individuals may conceal their sexual identity to avoid experiencing sexual prejudice and discrimination or they may perceive the LGBT community as unsupportive and thus avoid disclosing their sexual orientation or engaging with the LGBT community (Pachankis et al., 2015;Pereira & Leal, 2005). A recent meta-analysis reported that concealing one's sexual orientation is associated with mental health difficulties, particularly, internalizing problems such as psychological distress, depression, or anxiety (Pachankis et al., 2020). This association is partially explained by self-stigma and discrimination experiences that may lead to a chronic hypervigilance to avoid further negative experiences (Pachankis et al., 2008). Studies also show that experiencing or anticipating rejection after disclosure may reinforce self-stigma and difficulties accepting one's sexual orientation, which in turn may lead sexual minority individuals to conceal their sexual orientation to shield themselves from (further) negative experiences (e.g. Kelleher, 2009). While it is well known that disclosing one's sexual orientation has medium and long-term benefits, there are people who choose to conceal their sexual orientation from others, which can be protective against the stress of discrimination and violence. In such cases, concealing one's sexual orientation has been associated with lesser externalizing problems, namely, problematic substance use (Pachankis et al., 2020).
In Portugal, older sexual minorities have greater difficulties in disclosing their sexual orientation and may take longer to do so, particularly due to a repressive and persecutory social and political context that the older generation endured, whereas younger generations benefit from increasing social and legal acceptance and thus feel more empowered to disclose their sexual orientation earlier in their life course (Pereira & Monteiro, 2016). Further, Portuguese studies with older gay and bisexual men suggested that a lengthier disclosure gap may have negative consequences for one's relationships and general well-being Ribeiro-Gonçalves et al., 2019).
The middle-aged cohort may have experienced increased levels of psychological distress as they developed an awareness of their same-sex attraction as being dangerous and linked to the possibility of infection and death (Hammack et al., 2018;Harper et al., 2015;Meyer et al., 2021). Thus, not being able to explore their sexuality freely due to the constrains associated with the fear of HIV may have led individuals to have a more difficult time coming to terms with their sexuality (Hammack et al., 2018). The decades of 1980s and 1990s were also marked by strong social stigma, not only because of the association of gay male sexuality with fast physical health deterioration but also due to a resuscitated antigay religious speech claiming AIDS was a divine punishment, thus representing a new context for internalizing stigma and shame about one's sexuality (Hammack et al., 2018). In addition, middle-aged sexual minority men from Portugal did not have any formal support systems or LGBT networks. Much like the older cohort, middle-aged sexual minority men lacked access to accurate information about issues related to sexuality, formal LGBT resources and communities, which likely enhanced their feelings of alienation and isolation and forced them into heterosexual lives (Afonso, 2019;Mufioz-Plaza et al., 2002;Santos, 2018).
In contrast, younger sexual minority men in the adult cohort benefitted from a new cultural narrative on homosexuality as a legitimate expression of human diversity, at a time in which society started to shift toward embracing gay identity as a legitimate and immutable trait and away from pathologizing and demonizing homosexuality (Hammack et al., 2018). This growing acceptance of same-sex sexual orientation among the adult cohort has been supported by numerous studies, and these sexual minorities' coming of age is taking place in a socio-political environment far more differentiated than any generation before them (Frost et al., 2015;Frost et al., 2020). In the 1990s, the Internet emerged as a new social context that facilitated interactions among gay and bisexual men in numerous new ways (Harper et al., 2015). The Internet allowed gay and bisexual adolescents to explore their sexuality in virtual environments, thus being able to control when, how, and how much they wanted to disclose their sexual orientation (Harper et al., 2015). It is also important to note that this younger cohort came of age in a time where much of the basic LGBT rights were being reclaimed and conquered and the LGBT movement was flourishing throughout the western world and particularly in Portugal (e.g. Brandão, 2008). However, they still experience significant social and psychological challenges and may still feel the need to conceal their sexual orientation to protect themselves from social stigma insofar as prejudice and inequalities persist (Meyer, 2003;Meyer et al., 2021). Further, studies show that sexual minorities in this age cohort also have poorer mental health outcomes than their heterosexual counterparts (e.g. Gomes et al., 2020;Rosario et al., 2006), which might be explained by prevalent stigma, having to deal with both this new accepting reality and the old and more stigmatizing one in their coming of age (Frost et al., 2015).
The disclosure gap can be a time of great emotional instability and insecurity, which may have significant negative consequences for sexual minorities' well-being (Guittar & Rayburn, 2015). Older sexual minorities show worrisome levels of psychological distress, associated with concealing their sexual orientation from others (e.g. Ribeiro- Gonçalves et al., 2019). In addition, high levels of self-stigma negatively impact older sexual minorities' mental health . Although these high levels of self-stigma may be associated with the cumulative effects of minority stress throughout the life course, younger generations also show significant levels of self-stigma and psychological distress (Gomes et al., 2020;Herek, 2007;Meyer et al., 2021;Rosario et al., 2006).
In contrast, it has been established that relational, family, and community support promote greater selfacceptance, greater likelihood of self-disclosure, and a better psychosocial adjustment (Frost & Meyer, 2012;Oliveira et al., 2010). Studies show that the feeling of social isolation and low self-esteem resulting from stigma can be alleviated by family acceptance of sexual orientation/identity, satisfaction with friendship and romantic relationships, connectedness with the LGBT community, and general social support (e.g. Pereira & Costa, 2016). Among older sexual minorities, more proximal support (e.g. from a partner or family members) has been shown to foster a better management of the effects of stigmatization and social discrimination (Pereira et al., 2019). However, disengagement from the LGBT community may hinder them from enjoying the benefits associated with community identification and support insofar as community connectedness represents an opportunity for solidarity and group cohesion through shared experiences and goals. Further, it might constitute an important protective factor against minority stress, feelings of loneliness, and isolation (Díaz et al., 2004). Younger generations, particularly those in the adult cohort, seem to use these social resources more often than older generations, as they have only been available for the last two decades (Pereira & Monteiro, 2016;Ribeiro-Gonçalves et al., 2020).
Although there is a large bulk of studies addressing the protective and stressor.
psychosocial variables for sexual minorities, most studies focus only on one age cohort (e.g. gay and bisexual youth) and do not account for differences across generations of sexual minority men (e.g. Baiocco et al., 2016). The present study aims to examine stressor (i.e. selfstigma, psychological distress) and protective psychosocial factors (i.e. LGBT community connectedness) for the disclosure gap through a comparative retrospective cohort study, with three generations of sexual minority men: adults (25-39 years), middle-aged (40-59 years), and seniors (60 or more years).

Participants
This study is part of a larger research project about the experiences of stigma and mental and physical health of sexual minorities in Portugal, and the data was collected between 2018 and 2019. A sample of 274 cisgender men was recruited for this study, from which 231 (84%) selfidentified as gay and 43 (16%) self-identified as plurisexual (e.g. bisexual, pansexual). Sexual orientation was measured through an open-ended question, and responses were coded a posteriori. Participants' ages ranged from 25 to 79 years, with a mean age of 49 (SD = 14). Participants were divided into three age groups to allow comparisons between adult (25-39 years; n = 86), middle-aged (40-59 years; n = 88), and older men (60-79 years; n = 100). Most men were single (60%), approximately 43% reported living alone, and about one quarter had at least one child. Most men lived in urban areas (71%) and were full-time employed (54%), and close to one third had completed high school, 30% had completed a college degree, and 31% had completed a master or doctoral degree. Detailed sociodemographic data are presented in Table 1. Sexual minority men were compared on all sociodemographic characteristics, and only significantly differed on relational status in that plurisexual men were more likely than gay men to be married, χ 2 (1) = 6.757, p = 0.013 or in a committed relationship, χ 2 (1) = 7.808, p = 0.006. Considering the lack of overall differences between the two groups, the complete sample was used.

Measures
In addition to the sociodemographic questionnaire, the participants were asked to complete other standardized scales and grids.

Disclosure and acceptance of sexual orientation
To measure the level of disclosure and perception of acceptance of one's sexual orientation, participants were asked (1) how old they were when they first became aware of their sexual orientation (response option in years); (2) how old they were when they first disclosed their sexual orientation to others (response option in years; for participants who had not disclosed, the missing value was replaced for their current age); and (3) if they had ever disclosed their sexual orientation to anyone (response format: yes/no). The time elapsed between participants realizing they were sexual minorities and disclosing it to others was calculated and computed to create the variable (4) disclosure gap of sexual orientation.
To measure the level of disclosure of sexual orientation and perception of acceptance of sexual orientation, the Index of Disclosure of Sexual Orientation was used (Costa et al., 2013). This index is divided into two questions: 1. "Who knows about your sexual orientation?" assessed with four possible answers: (i) knows it and we have talked about it; (ii) knows or suspects it but we never talked about it; (iii) doesn't know it, and (iv) it doesn't apply. 2. "How do you feel about their acceptance of your sexual orientation?" assessed with five possible answers: (i) accepts it very well; (ii) with some difficulty in the beginning but now accepts it well; (iii) (still) have some difficulty with it; (iv) doesn't accept it; and (v) it doesn't apply. Participants were then asked to refer their answers to parents, siblings, friends, and peers/colleagues, separately. Responses were measured as ordinal variables in which higher scores reflected lower disclosure level and lower acceptance of sexual orientation.

Self-stigma
Self-stigma was measured using the Questionnaire of Homosexual Identity . This questionnaire measures feelings and beliefs about one's sexual orientation, and it is composed by 13 items measured in a 5-point Likert scale (from 1-completely disagree to 5-completely agree). A confirmatory factor analysis corroborated the one-dimensional structure after the deletion of one item ("I feel that my sexual orientation is just one characteristic of my identity", λ = 0.105, p = 0.099), χ 2 (48) = 124.098, CFI = 0.957, GFI = 0.927, RMSEA = 0.076 95% [0.060, 0.093]. Mean scores were computed so that higher scores reflected higher levels of selfstigma. Cronbach's alpha for this study was high (α = 0.907).

Psychological distress
Psychological distress was measured using the Kessler's K6 Psychological Distress Scale (Kessler et al., 2002), composed by 6 items measured in a 5-point Likert scale (from 0, completely disagree, to 4, completely agree). Mean scores were computed so that higher scores reflected higher levels of non-specific psychological distress. Cronbach's alpha for this study was high (α = 0.912).
LGBT community connectedness Community connectedness was measured using the Connectedness to the LGBT Community Scale (Frost & Meyer, 2012), composed by 8 items measured in a 4-point scale from (from 1, completely disagree, to 4, completely agree). Mean scores were computed so that higher scores reflected greater involvement with the LGBT community. Cronbach's alpha for this study was high (α = 0.909). Participants were further asked how often they interacted with the LGBT community, which was measured on a 4-point scale from (1, never, to 4, often), with higher scores reflecting greater interaction with the LGBT community.

Procedures
Convenience and intentional non-probabilistic sampling were used. The study's survey was advertised through different electronic means and social networks, namely, Facebook groups, blogs, and associations directed to the LGBT community. Inclusion criteria for this study were (1) being 25 years old or older; (2) identifying as a sexual minority; (3) identifying as men (gender identity); and (4) being assigned male at birth (sex). Potential participants were given a brief description of the study and a link to an online survey hosted on Qualtrics. Informed consent was requested on the first page of the survey, before participants completed any measures. All procedures were in accordance with the ethical standards of the Portuguese Psychological Association and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Disclosure and Acceptance of Sexual Orientation
Most cisgender men in this sample (80%) reported having disclosed their sexual orientation to at least one person. When asked about when they first realized they were sexual minorities, responses varied between 4 and 59 years of age (M = 17.03; SD = 7.83). When asked about when they first disclosed their sexual orientation, responses varied between 12 and 76 years of age (M = 30.75; SD = 15.81). Regarding the time elapsed between self-realization and disclosure of sexual orientation (disclosure gap), responses ranged from 0 to 58 years (M = 13.71; SD = 14.64).
In order to compare the experiences of disclosure of sexual orientation among the three age cohorts, a one-way ANCOVA was conducted while controlling for sexual orientation effects (gay/plurisexual). The effects of sexual orientation were significant for age at realization of sexual orientation, F(2, 268) = 16.390, p < 0.001, η p 2 = 0.058; age at disclosure, F(2, 268) = 18.437, p < 0.001, η p 2 = 0.065.; and disclosure gap, F(2, 268) = 3.819, p = 0.050, η p 2 = 0.014. Plurisexual men took significantly longer than gay men to realize their sexual orientation (M = 16.23 and M = 21.37, respectively), to disclose their sexual orientation (M = 29.00 and M = 40.19, respectively), and had a larger disclosure gap (M = 12.81 and M = 18.57, respectively), although the effect sizes for these differences were small.
After controlling for sexual orientation effects, significant differences between the three age cohorts were found for age at disclosure, F(2, 268) = 52.691, p < 0.001, η p 2 = 0.285., and disclosure gap, F(2, 268) = 41.736, p < 0.001, η p 2 = 0.240 (Table 2). No significant differences were found for age at realization of sexual orientation, F(2, 268) = 2.575, p = 0.078, η p 2 = 0.019. As shown in Fig. 1, age at realization, age at disclosure, and disclosure gap all significantly increased with age insofar as older men realized their sexual orientation, disclosed their sexual orientation, and took a significantly longer time to disclose it than their middle-aged counterparts, who, in turn, took a significantly longer time than their younger counterparts.
When asked about who knew about participants' sexual orientation, close to 50% reported having discussed it with parents, 51% with siblings, 57% with friends, and 39% with peers/co-workers. When asked about the reactions to the disclosure, 61% reported feeling accepted by parents, 71% by siblings, 81% by friends, and 69% by peers/co-workers. In order to compare the age cohorts on levels of disclosure of sexual orientation and perception of acceptance of sexual orientation, non-parametric Kruskal-Wallis tests were performed as indicated for ordinal variables (Table 3). Test results were significant for disclosure to parents, H(2) = 28.704, p < 0.001; siblings, H(2) = 22.063, p < 0.001; friends, H(2) = 74.419, p < 0.001; and peers/co-workers, H(3) = 26.972, p < 0.001. Significant differences were also found for perception of acceptance by parents, H(2) = 16.093, p < 0.001; siblings, H(2) = 41.253, p < 0.001; friends, H(2) = 62.296, p < 0.001; and peers/co-workers, H(3) = 41.658, p < 0.001. To assess these significant differences further, Rank Cases were performed followed by one-way ANOVAs using the ranked variables and Tukey post hoc tests for multiple comparisons. Overall, both the level of disclosure and the perception of acceptance significantly differed in all pairwise comparisons regarding parents, siblings, friends, and co-workers (all p's < 0.05). Older men tended to disclose their sexual orientation less than middle-aged and adults and tended to feel less accepted following disclosure than middle-aged and adults. A similar pattern was found when comparing middle-aged with adult men.

Self-Stigma, LGBT Community Connectedness, and Psychological Distress
For self-stigma, the mean score for the whole sample was 2.43, suggesting moderate to low levels of self-stigma. When comparing the three age cohorts, significant differences were found, F(2,273) = 10.893, p < 0.001, η p 2 = 0.074 (Table 4). Pairwise comparisons through Tukey post hoc tests indicated that older men showed significantly higher levels of selfstigma than middle-aged and adult men.
For LGBT community connectedness, 15% of participants stated that they never have any interaction with the LGBT community. The mean score for the level of LGBT community connectedness in this sample was 2.80, indicating a low level of involvement with the LGBT community. Significant differences between the three age cohorts were found, F(2,273) = 5.932, p = 0.003, η p 2 = 0.042, in that the adult cohort reported significantly greater community connectedness than middle-aged and older men.
Regarding levels of psychological distress, the responses ranged from 0 to 24, with a mean score of 7.08. Kessler et al. (2002) proposed a cut-off score equal to or above 13 as indicative of serious mental suffering, while recent studies (e.g. Prochaska et al., 2012) proposed that a score equal to or above 5 indicates low to moderate levels of distress. In the current sample, 41% of the men scored below 5, 40% scored between 5 and 12, and 19% scored 13 or over. Both the mean score and the percentage of cisgender men who scored over  1 Mean years for age at realization, age at disclosure, and gap years between age at realization and age at disclosure of sexual orientation in the three age cohorts the threshold of 5 suggest a high prevalence of moderate to severe psychological distress among this sample. Although the adult cohort scored higher on psychological distress than the other two groups, no significant differences were found, F(2,273) = 2.012, p = 0.136, η p 2 = 0.015.

Predicting the Disclosure Gap
Individual stepwise multiple linear regressions were conducted for each age cohort. As shown in Table 5, none of the variables significantly contributed to the explained variance of disclosure gap of the younger cohort. For middle-aged men, only self-stigma emerged as significant (R 2 adj = 0.105), indicating that higher level of self-stigma explained a larger disclosure gap. For older men, only LGBT community connectedness emerged as significant in explaining the disclosure gap (R 2 adj = 0.041), indicating that lower levels of LGBT community connectedness explained a larger disclosure gap.

Discussion
This study set out to examine the coming out experiences, including age of self-awareness, age of disclosure of sexual orientation, and disclosure gap between self-awareness and disclosing one's sexual orientation and to evaluate the role of self-stigma, LGBT community connectedness, and psychological distress in explaining the disclosure gap in three age cohorts of sexual minority men. The overwhelming majority of the men in this study had disclosed their sexual orientation to at least one person although there were wide variations in the timing of awareness, disclosure, and disclosure gap across the three age cohorts. Further, the experiences regarding the people to whom these men had disclosed and how accepted they felt after disclosure also varied as a function of age cohort. As expected, the findings confirmed that the overall coming out experiences worsened with age, with the older cohort (i.e. men aged 60 and over) reporting, realizing and disclosing their sexual orientation later in life, taking longer to disclose it to others and feeling less accepted by significant others following disclosure.
Several studies show that sexual minority men are disclosing their sexual orientation earlier, and this trend may be due to looser constrains imposed by social stigma when compared to how it was before the turn of century (e.g. Grov et al., 2006;Savin-Williams & Cohen, 2015). Thus, it is not surprising that the younger age cohort in this study realized and disclosed their sexual orientation earlier than their older counterparts from the other two cohorts. The Internet has been used by sexual minority men as a tool for gaining accurate information about sexual minority experiences and identities and connecting with similar others, thus providing them a greater sense of comfort and acceptance in which to explore and affirm their sexuality (e.g. Harper et al., 2015). Their access to the virtual world in a critical developmental moment (i.e. adolescence) might have considerably eased their their coming out experiences when compared to the other two older age cohorts. In addition, the development of social policies, formal social support and LGBT networks, and legal changes aimed at protecting sexual minorities has increased significantly and benefited the younger cohorts more (Grov et al., 2006;Hammack et al., 2018;Pereira & Monteiro, 2016). These legal changes most likely had a greater impact for the two younger cohorts, especially adults, as they are more likely to make use of them (e.g. marriage and parenting) than seniors (Pereira & Monteiro, 2016). In contrast, older sexual minority men reported realizing and disclosing their sexual orientation much later than their younger counterparts, a wider disclosure gap, and feeling less accepted after disclosure to significant others (parents, siblings, friends, and peers/colleagues). In spite of the present sociohistorical moment that protects and affirms sexual minorities' human rights, older sexual minority men spent most of their life course concealing their sexual orientation in order to protect themselves from harassment, persecution, and stigma. In Portugal, the older cohort came of age in a time of a highly repressive and persecutory dictatorship, during which homosexuality was criminalized. In this context, this generation developed their sexual orientation in forced hiding, many have failed to achieve important developmental milestones, and many have entered a heterosexual marriage to protect themselves from negative social consequences (Afonso, 2019;Almeida, 1995). This context also conditioned the expectations and reactions from support networks, and the assumption of the sexual minority orientation can be seen as more drastic, rarer, and less expected, sometimes creating greater resistance and rejection perceived and/or felt by people who identify as such (Dunlap, 2016;Pistella et al., 2016). Important cohort differences were found in explaining the disclosure gap. For adults, none of the examined variables (self-stigma, LGBT community connectedness and psychological distress) significantly explained the disclosure gap, possibly because the disclosure gap was small and the disclosure had happened at a very young age. Nevertheless, younger generations of sexual minority men still demonstrate concerning levels of psychological distress, as demonstrated in previous studies (e.g. Meyer et al., 2021), including in Portuguese studies (e.g. Gomes et al., 2020). In this study, the younger cohort scored higher than the other two age cohorts on psychological distress, although these differences did not reach significancy. This finding may seem counterintuitive, but it is not entirely novel. The younger generation might be more empowered to disclose their sexual orientation and initiate the process of integrating their sexual minority identity earlier in life (Savin-Williams & Cohen, 2015). Greater social contact with other sexual minority individuals has been shown to help come to terms with one's sexual orientation, as well as to provide support to cope with potential barriers to coming out (Padilla et al., 2007). Nevertheless, visibility may be also a disadvantage as younger sexual minority men may be more exposed to chronic stress, prejudice, and victimization even before they reach adulthood. The constant public and political discussions around LGBT rights can create sources of minority stress and a harmful environment for sexual minorities, particularly for those who are more publicly out and engaged with the LGBT community (Meyer et al., 2021). US studies have shown that public discussions about LGBT rights that accompany policy initiatives (e.g. same-sex marriage) can have a negative impact for LGBT people, especially if these policies are not enacted (for a comprehensive discussion, see Costa, 2021).
For middle-aged sexual minority men, only self-stigma significantly explained the disclosure gap in that for this generation negative feelings about one's sexual orientation seems contribute to a delayed disclosure. Considering that the men in the middle-aged cohort had their adolescence/ young adulthood marked by the rise of the AIDS pandemic, they most likely experienced increased levels of social stigma and developed an awareness of same-sex attractions as potentially dangerous. As such, exploring their sexuality was embedded in stigma and fear of acquiring AIDS, which contributed for internalizing stigma and shame about their same-sex attractions (Hammack et al., 2018). In addition, sexual minority men from Portugal could not enjoy some of the achievements that were happening in the USA to counterbalance some of the negative effects (e.g. LGBT organizations), and much like the generation before them, they were being led into heterosexual lifestyles to avoid persecution, violence, or criminalization of their same-sex behaviour (Afonso, 2019;Cascais, 2006;Santos, 2018).
Studies also show that sexual minorities from stigmatizing contexts present a greater risk of internalizing negative feelings toward themselves, leading them to encounter significant difficulties and psychological barriers in disclosing their sexual orientation (e.g. Pistella et al., 2016). For older sexual minority men, only a weak engagement with the LGBT community significantly explained taking longer to disclose their sexual orientation or not disclosing it at all. This finding may be explained by the lack of support from others with whom to identify and share experiences, possibly associated with a heteronormative family experience, which makes it harder to disclose (Williams & Fredriksen-Goldsen, 2014). This lack of social and community support can also increase the likelihood of feelings of loneliness, isolation, and invisibility. The connection to the LGBT community is generally weaker among the elderly who did not have an established community (e.g. community organizations) for much of their life.
Lastly, irrespective of the generational cohort, the sexual minority men in this study reported generally low levels of self-stigma but a significant lack of engagement with the LGBT community and worrisome levels of psychological distress. Studies show that across different generations of Portuguese sexual minorities, there are worrisome levels of distress (Gomes et al., 2020;Ribeiro-Gonçalves et al., 2019), which makes them more prone to mental health difficulties and poorer physical health. On the one hand, stigmatization and discrimination experiences have been strongly associated with higher levels of psychological distress (e.g. Kelleher, 2009) which might explain why even though the men in this study reported low self-stigma, they still showed moderate to high levels of psychological distress. On the other hand, the finding of a weak engagement with the LGBT community and high levels of psychological distress among the three cohorts of sexual minority men is not surprising since studies show that community connectedness represents an important protective factor against minority stress (e.g. Díaz et al., 2004).

Limitations and Strengths
This study had some limitations that warrant acknowledgement. Firstly, the participants were recruited through convenient and intentional non-probabilistic sampling, which may have compromised the representativeness of this sample. Further, although common in the field, online sampling is known to possibly bias samples by favouring middle to high class participants, with higher formal education, and with greater technological skills; this is particularly relevant in the case of older men who generally have low technological and Internet skills. Moreover, this study used a cross-sectional retrospective design so causality cannot be ascertained, and the findings ought to be taken with caution regarding the experiences of older and middle-aged men whose reports may have been influenced by their current context and situation. This potential retrospective bias is particularly important when reading the associations between disclosure gap and self-stigma, LGBT community connectedness, and psychological distress, as these reflect current experiences and do not necessarily indicate causal relations between them and the disclosure gap.
The disclosure gap was measured by asking participants two separate questions, which constitutes an objective measure and a methodological strength. However, we did not specifically assess to whom they first disclosed and whether they disclosed their sexual orientation one or multiple times. Further studies could clarify the more subtle age cohort differences by examining multiple disclosures in greater depth. Outside of the scope of this study albeit extremely relevant would be to examine how participants make sense of the disclosure gap in their own words or, more specifically, what factors do different generations of men identify as responsible for delaying or preventing their developmental milestones in affirming their sexual identity. Lastly, in this study, we have only addressed the coming out experiences of sexual minority men. Future studies could further investigate differences between sexual minority men and women across different generations.
Despite these limitations, this study also had some strengths. Specific to the context of the study, to the best of our knowledge, this is the first study conducted in Portugal examining the coming out experiences of sexual minority men across different age cohorts. Further, both in Portugal and elsewhere, most studies focus only on one age cohort and do not account for differences across generations. This study served to illuminate some of the different experiences between adults, middle-aged, and older sexual minority men. In addition, this study highlighted some of the social factors that may improve sexual minority's psychosocial well-being and possibly counterbalance the negative effects of stigma, namely, connection to the LGBT community. Studies that identify and examine factors that may ameliorate the negative consequences of minority stress across the life cycle are much needed, especially among the most vulnerable such as older sexual minority men.

Availability of Data and Material
The raw data supporting the conclusions of this article may be made available by the authors.

Declarations
Competing Interests The authors declare no competing interests.