The Journal of Primary Prevention

, Volume 31, Issue 5, pp 273–309

A Systematic Review of Parental Influences on the Health and Well-Being of Lesbian, Gay, and Bisexual Youth: Time for a New Public Health Research and Practice Agenda

  • Alida Bouris
  • Vincent Guilamo-Ramos
  • Angela Pickard
  • Chengshi Shiu
  • Penny S. Loosier
  • Patricia Dittus
  • Kari Gloppen
  • J. Michael Waldmiller
Literature Review

DOI: 10.1007/s10935-010-0229-1

Cite this article as:
Bouris, A., Guilamo-Ramos, V., Pickard, A. et al. J Primary Prevent (2010) 31: 273. doi:10.1007/s10935-010-0229-1

Abstract

Relatively little is known about how parents influence the health and well-being of lesbian, gay, and bisexual (LGB) adolescents and young adults. This gap has led to a paucity of parent-based interventions for LGB young people. A systematic literature review on parental influences on the health of LGB youth was conducted to better understand how to develop a focused program of applied public health research. Five specific areas of health among LGB young people aged 10–24 years old were examined: (a) sexual behavior; (b) substance use; (c) violence and victimization; (d) mental health; and (e) suicide. A total of 31 quantitative articles were reviewed, the majority of which were cross-sectional and relied on convenience samples. Results indicated a trend to focus on negative, and not positive, parental influences. Other gaps included a dearth of research on sexual behavior, substance use, and violence/victimization; limited research on ethnic minority youth and on parental influences identified as important in the broader prevention science literature; and no studies reporting parent perspectives. The review highlights the need for future research on how parents can be supported to promote the health of LGB youth. Recommendations for strengthening the research base are provided.

Keywords

Parental influences Gay Lesbian Bisexual Same-sex attraction Adolescents Young adults Health risk behavior Sexual behavior Mental health Violence Victimization Suicide Substance use 

Introduction

Research on the health and well-being of lesbian, gay, and bisexual (LGB) adolescents and young adults (i.e., youth) has grown considerably in the past 20 years (Anhalt and Morris 1998; Maher et al. 2010). In part, the increase has reflected concern over data indicating that LGB youth are a vulnerable group whose health warrants a significant public health focus. For example, studies indicate that many LGB youth encounter stigma, verbal harassment, or physical violence in response to their sexual orientation (Bontempo and D’Augelli 2002; Huebner et al. 2004). Experiences with victimization and discrimination have been associated with mental health distress (Hershberger and D’Augelli 1995), and research has shown that, compared to heterosexual youth, LGB youth are at an increased risk for suicide (Russell and Joyner 2001; Silenzio et al. 2007) and report elevated rates of sexual risk behavior and substance use (Garofalo et al. 1998; Marshall et al. 2008). In addition, epidemiological data show that young men who have sex with men, especially young men of color, continue to be disproportionately affected by HIV/AIDS (Campsmith et al. 2008; Hall et al. 2008).

In response to this growing body of evidence, prevention scientists have sought to identify mechanisms that can be targeted to support LGB youths’ health and well-being. In general, this research has tended to focus on the identification of risk factors, with limited attention to protective factors that may support resilience in the lives of LGB young people (Savin-Williams 2001). In addition, studies with LGB youth have tended to neglect key developmental contexts that have been identified as important correlates of youth health in the broader prevention science literature. For example, there now exists a large body of research on the importance of parents for understanding the health and well-being of young people (Steinberg 2001; Wood et al. 2004). Both the National Institutes Health and the Centers for Disease Control and Prevention (CDC) have identified parents as a critical site for promoting young people’s health (CDC 2006; Pequegnat and Szapocznik 2000), and numerous parent-based interventions have been developed and evaluated with promising results (Guilamo-Ramos et al. 2010; Stanton et al. 2004; Turrisi et al. 2001).

In contrast, the majority of research on LGB youth has not included similar parental approaches (Garofalo et al. 2008). To date, most research has examined the context, process, and outcomes associated with parents’ responses to learning about a child’s sexual orientation (for reviews, see Heatherington and Lavner 2008; Savin-Williams 1998), with limited attention to the dimensions of parenting commonly studied in prevention science research with heterosexual youth. This paucity of research is striking given that parents and families are widely recognized as one of the most salient developmental contexts for young people (Collins et al. 2000; Steinberg 2001). Developmental theorists agree that parents can influence children through numerous mechanisms, including the transmission of parental values and expectations, role modeling, external reinforcement, parenting style, and the use of different parenting practices (Bandura 1975; Kasser et al. 1995; Steinberg 2001). Although a comprehensive review of this literature is beyond the scope of the present review, we briefly discuss key findings below.

Research on the role of parents in shaping youth health has tended to focus on two potential mechanisms of influence: (a) parenting style and (b) parenting practices. Although sometimes used interchangeably, important differences exist between the two (Darling and Steinberg 1993; Mize et al. 1998). Parenting style encapsulates the overall emotional climate between a parent and child and is characterized by the extent to which parents exhibit warmth and control in the parent–child relationship (Mize et al. 1998). Initially, three parenting styles were identified: (a) authoritative, characterized by a balance between control and warmth; (b) authoritarian, featuring high levels of parental psychological control and low levels of parental warmth; and (c) permissive, in which parents give children too much freedom and too little supervision (Baumrind 1971). Subsequent research has since divided permissive parenting into two distinct styles: (a) indulgent and (b) neglectful (Maccoby and Martin 1983). Across research studies with diverse youth, children of authoritative parents tend to engage in less health risk behaviors and report better psychosocial outcomes than children of authoritarian or permissive parents (Montgomery et al. 2008; Pettit et al. 2001; Steinberg 2001).

In contrast to the more global nature of parenting style, parenting practices refer to the specific and goal-oriented strategies that parents employ as they raise their child (Mize et al. 1998). These include practices such as parent–child communication about sex, as when a father talks with his young adult son about why he should use a condom during sex with the goal of helping his son avoid HIV, or parental monitoring of an adolescent’s friends, so that a mother can buffer her child from potentially negative peer influences. Parenting practices identified as important correlates for young people’s health include (a) parent–child connectedness and the establishment of a close parent–child relationship (Markham et al. 2010); (b) parent’s psychological and behavioral control (Barber et al. 2005); (c) parental support (Barber et al. 2005); (d) parental attitudes and values that discourage risk taking, such as parental disapproval or the endorsement of educational and career goals (Ford et al. 2005; Frisco 2005); (e) parent–child communication (Guilamo-Ramos et al. 2007); and (f) parental monitoring and supervision (Stattin and Kerr 2000).

In general, research has found that higher levels of positive parenting practices are inversely associated with health risk behaviors and maladaptive health outcomes (Markham et al. 2010). For example, studies show that parenting practices such as establishing a close, involved, and loving parent–child relationship is associated lower levels of youth suicidality (Flouri 2005). Although most research examines direct effects, both moderated and mediated relationships have been observed (Turrisi et al. 2007), as have non-linear relationships (Barber et al. 2005). Research also has noted that parenting and the parent–child relationship are dynamic processes that change over time. For their part, parents use different practices for sons and daughters, for older and younger children, and in response to a child’s unique characteristics and needs (Crouter and Booth 2003). In return, youth often engage in specific behaviors, such as open or selective disclosure of information about their life, that seek to influence how their parents respond to and parent them (Crouter and Booth 2003). Finally, studies also show that parents can remain important, even as peers, school, and work become increasingly salient (Turrisi et al. 2001; Wood et al. 2004).

As the evidence base on the protective role of parents has grown, so too has support for interventions that target parents as a primary mechanism through which to improve youth health. There now are promising parent-based interventions for sexual risk behavior (for a review, see Robin et al. 2004), smoking (for a review, see Thomas et al. 2007), drug and alcohol use (Kumpfer and Alvarado 2003; Turrisi et al. 2001), mental health (for a review, see Hoagwood 2005), and violence (Kumpfer and Alvarado 2003). Typically, these interventions seek to strengthen specific parenting practices, such as the nature and quality of a parent’s monitoring efforts (Stanton et al. 2004). Although there are differences across interventions, all aim to provide parents with the knowledge, skills, and support necessary to support their child’s health, and they all capitalize on the parent–child relationship as a primary mechanism through which to effect change.

In comparison, we know far less about parenting in families with an LGB child. In general, research with LGB youth has tended to reflect strained parent–child relationships (Darby-Mullins and Murdock 2007), parental rejection of youth (Ryan et al. 2009), or an assumption that parents are not positively involved in their child’s life (Garofalo et al. 2008). Although these experiences are true for some LGB youth, not all LGB young people experience parental rejection and hostile family environments (Garofalo et al. 2008). Indeed, emerging research indicates that many parents are aware of their child’s unique needs (LaSala 2007), are open to understanding how their responses influence their child’s well-being (Ryan et al. 2009), and want assistance in supporting their child’s health (LaSala 2007). Despite these factors, there are currently no known parent-based interventions that have been developed to help parents support the health and well-being of their LGB child.

The lack of research in this area represents a critical gap in the prevention science literature as LGB young people are similar to their heterosexual peers in many ways—they date and form romantic relationships, they develop friendships and interact with their peers, they attend school and plan for their future, and they encounter situations that present opportunities for engaging in risk behaviors (Bauermeister et al. 2010; Savin-Williams 2001). In these contexts, it is plausible that LGB youth may benefit from their parents’ efforts in ways that are similar to their heterosexual peers, especially when these practices occur in the context of a good parent–child relationship and are tailored for needs of LGB youth. At the same time, LGB youth are different from their heterosexual peers in that they often explore their sexual orientation in settings that lack formal guidance or may be dangerous for their mental and physical health (Bontempo and D’Augelli 2002). Experiences with homophobia have implications not only for youth but also for parents as they are tasked with responding to their LGB child as he or she moves through the coming out process. Here, parents can play two particularly important roles. First, they can serve as an important source of support by accepting their child unconditionally and by buffering them from negative and harmful responses to their sexual orientation. They can improve their own knowledge base on the needs of LGB youth and make positive adaptations to their parenting practices to reflect this new information (D’Augelli 2005). Alternatively, parents may serve as a source of stress as they express ambivalence, intolerance, or rejection after learning of their child’s sexual orientation and, in some cases, perpetrate sexual orientation-related victimization against their LGB child (D’Augelli 2005).

In this regard, there are two interrelated dimensions to parenting LGB youth: (a) one that focuses on the types of parenting practices traditionally studied in the prevention science literature, such as parent–child connectedness, communication, or monitoring, and (b) one that focuses on the unique aspects of parenting an LGB child, namely, parents’ responses to their child’s sexual orientation and the extent to which parents serve as a source of support or stress in this domain. Currently, we know relatively little about how these domains of parenting work together as theoretical frameworks of parenting that bring these dimensions together are rare. To date, we know of only one conceptual framework that has attempted to integrate broader parent–child interactions with those specific to having an LGB child (see Heatherington and Lavner 2008), and it has not yet been evaluated in empirical research.

As research on LGB youth continues to grow, researchers are calling for greater attention to the contextual influences that shape their health and well-being, especially parents and family systems (D’Augelli 2005; Garofalo et al. 2008; Horn et al. 2009), and for the development of parent-based interventions to support LGB youth’s health (Garofalo et al. 2008). The overall purpose of the present review was to assess the current state of knowledge on parental influences on the health and well-being of LGB adolescents and young adults in order to develop a set of focused recommendations for a parent-based research agenda. In doing so, we sought to understand the nature of parental influences on the health of LGB youth, assess how parental influences are studied in relation to the larger body of empirical research on heterosexual youth, identify key gaps in the literature, and support prevention scientists to begin the process of developing parent-based interventions for LGB youth. The review targets five specific areas of LGB health: (a) sexual risk behaviors and related health outcomes, e.g., sexually transmitted infections (STIs) and HIV; (b) experiences with violence and victimization; (c) substance use including alcohol, tobacco, and other drug use; (d) suicide; and (e) mental health and well-being, including clinical indicators of mental health, such as depression or anxiety, and more general indicators of youth well-being, such as self-esteem. These areas were targeted because (a) they represent key health behaviors and outcomes for all young people, regardless of their sexual orientation; (b) research suggests that LGB youth experience health disparities in each domain; and (c) studies indicate that parents can play an important role in promoting youth health in each area (Resnick et al. 1997).

Method

Literature Search

Peer-reviewed articles examining parental influences on LGB youth health were identified via three primary methods. First, a computer-based search of the following databases was conducted: PubMed, PsycInfo, PsycArticles, the Institute for Scientific Information Web of Science, and Social Service Abstracts (a full list of the search terms is available from the second author). Second, an ancestral approach was used (White 1994), which entailed reviewing the reference lists of each included article to identify studies for possible inclusion. Finally, literature reviews on LGB youth and on adolescent and young adult health were examined in order to identify other potentially relevant articles.

Inclusion Criteria

Articles were included if they (a) were quantitative research published in a peer-reviewed journal between 1980 and 2010, (b) sampled U.S. youth primarily between the ages of 10–24 years old, and (c) examined parental influences as a correlate of one of the five targeted health areas. A broad age range was selected for several reasons. First, adolescents and young adults were targeted in order to ensure that a representative sample of the literature was obtained. This was particularly important given our interest in making applied recommendations. Although there are differences in the parent–child relationship across the ages of 10–24, research suggests that parents can influence their child’s health behaviors in early adulthood (Turrisi et al. 2001) and that parenting practices encountered in adolescence are associated with health indicators in early adulthood (e.g., Frisco 2005). Retrospective studies of older adults reporting on adolescent and young adult experiences were excluded due to the bias associated with long recall periods. Studies with transgender youth and with homeless and runaway LGB youth also were excluded as we hypothesized that both groups of young people experienced family and health concerns that were distinct from the larger population of LGB youth.

The term parents was operationalized to include biological, resident, non-resident, and step- and adoptive parents, as well as legal guardians and primary caregivers. Although the nature of parental influences may differ as a function of the particular composition of a parent–child dyad, the diversity of American families and the sparseness of the literature necessitated a broad definition of parents. In turn, the construct of parental influences was operationalized to encompass a range of parenting behaviors, including individual parenting practices, parenting style, and parent’s knowledge of and responses to a child’s sexual orientation. No criteria on the nature of parental influences were imposed, and studies with positive, negative, or no significant associations were included.

Results

A total of 31 articles met the study inclusion criteria. Of the 31 studies, 26 were cross-sectional, 2 were retrospective, and 3 were prospective. All three prospective studies utilized data from the National Longitudinal Study of Adolescent Health (Add Health). No experimental studies were located. Apart from the Add Health studies, only one other article utilized a probability sample (O’Donnell et al. 2002). Most studies obtained a waiver of parental consent when working with youth under age 18. Exceptions occurred when study data was drawn from nationally representative or school-based studies where LGB youth were not the primary focus of the research. None of the identified studies reported on parenting data collected from parents; thus, all results represent LGB youths’ perceptions of their parents. The sampling characteristics for each article are presented in Tables 1, 2, 3, 4, 5.
Table 1

Parental influences on LGB sexual behavior and health outcomes

Citation

Study design

Sampling strategy

Sample characteristics

Age

Parental influences

Sexual behavior and outcomes

Ackard et al. (2008)

Cross-sectional

2004 Minnesota Student Survey

Subsample of youth in grades 6, 9, and 12; excluded sexually inactive males

Passive parental consent (active if required by school or district)

Self-administered closed-ended survey during class period

“Saturated” sample: state-wide school-based sample of all youth in school on day of survey

88% of school districts participated: equivalent to 75% of 9th grade and 55% of 12th grade students in Minnesota

Refusal bias not reported

N = 10,095

100% male

82.9% White; 6.7% Black; 4.3% Asian; 6.1% Other/mixed

11.1% men who have sex with men and women; 1.5% men who have sex with men; 87.5% men who have sex with women

13–19

M = 16.7

SD = 1.5

Parent–family connectedness

Negatively associated with the number of sexual partners among all sexually active males who reported disordered eating, regardless of sexual orientation

Ford et al. (2005)

Prospective

Add Health: Wave I and III in-home interviews

Wave I: Parental consent and adolescent assent. Wave III: Young adult consent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

75.7% of Wave I respondents located and participated in Wave III

Refusal bias not reported

N = 14,332

50.8% male

49.2% female

67.6% White; 16% African American; 11.9% Latino; 3.7% Asian or Pacific Islander; 0.8% Native American

7.6% same-sex sexual attraction/ behavior; 92.4% heterosexual

18–28

Mean age not reported

18–20: 28.6%

21: 16.5%

22: 16.6%

23: 15.4%

24–28: 22.8%

 

Results control for same-sex sexual attraction/behaviors, demographic, school, individual, and other parent/family factors

Parent–family connectedness

Not associated with being diagnosed with STI

Parental disapproval of sex

For females, negatively associated with odds of being diagnosed with STI

For males, not associated with odds of being diagnosed with STI

Parental disapproval of adolescent contraception

Not associated with being diagnosed with an STI in early adulthood

Garofalo et al. (2008)

Cross-sectional

Subsample of YMSM from a larger survey of 496 LGBT youth: excluded females and transgender youth

Waiver of parental consent obtained; youth provided verbal

Computer-assisted self-administered interview

Convenience sample

Multiple recruitment strategies in Chicago: snowball sampling; email ads on high school and college list-serves; handed out palm cards in gay neighborhoods; flyers in retail areas and gay youth serving organizations

No recruitment from high-risk venues, e.g., bars, clubs or bathhouses

Participation rate and refusal bias not reported

N = 302

100% male

30% White; 33% Black; 26% Latino; 3% Asian;

8% Other / Multiracial

100% MSM

16–24 

M = 20.3

SD = 2.34

Parent–family connectedness

Negatively associated with odds of being HIV-positive, controlling for age and race/ethnicity

 

Not associated with odds of having had unprotected anal sex in past 12 months, controlling for age, race/ ethnicity and HIV serostatus

 

Not associated with having had multiple anal sex partners in past 3 months, controlling for age, race/ ethnicity and HIV serostatus

O’Donnell et al. (2002)

Cross-sectional

Baseline survey of Hermanos Jovenes, Community Intervention Trial for Youth

Subsample of youth reporting sexual activity in past 3 months

Interviewers obtained informed consent. Parental consent procedures not reported

Field staff interviewed respondents

Venue-based probability sampling of high and low attendance venues (bars, cafes, parks, LGB social service programs/events) in New York City: Bronx, Queens, and Washington Heights/Upper Manhattan

Bronx/Wash. Heights: 578 approached; 93% screened; 48% eligible; 99.6% enrolled

Queens: 637 approached; 89% screened; 46% eligible; 99% enrolled

N = 465

100% male

100% Latino; 40% foreign born

74% gay; 22% bisexual; 4% other

15–25

M = 21.4

SD = 2.5

 

All results control for demographic factors, gay self-identification, peer knowledge about MSM behavior, social support, and ethnic and gay community attachments

Parental knowledge of MSM behavior

Not associated with unprotected anal sex in past 3 months

Not associated with unprotected anal sex during last sex with a main partner

Not associated with unprotected anal sex at last sex with a non-main partner

Resnick et al. (1997)

Cross-sectional

Add Health: Wave I in-home interviews; excluded youth reporting sexual debut before age 11

Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

79.5% participation rate; refusal bias not reported

N = 11,572

Gender not reported

Race/ethnicity not reported: nationally representative sample

Sexual orientation not reported

7th–12th grade students

Mean age not reported

 

All results control for same-sex sexual attraction/behaviors, demographic factors, school factors, individual factors, and other parent and family factors:

Parent–family connectedness

Associated with older age of sexual debut for all 7th–12th grade youth

Parental disapproval of sex

Associated with older age of sexual debut for all 7th–12th grade youth

Parental disapproval of adolescent contraception

Associated with an older age of sexual debut for all 7th–12th grade youth

Ryan et al. (2009)

Retrospective

Family Acceptance Project

Subsample from larger study (N = 245); excluded transgender young adults (n = 21)

Consent procedures not reported

Computer assisted or pencil-and-paper self-administered surveys

Convenience sample

Recruited from 249 LGB venues located 100 miles of project office in urban city: 50% clubs/bars and 50% social service and community organizations

Participation rate and refusal bias not reported

N = 224

51% male;

49% female

48% non-Latino white; 52% Latino

Lesbian, gay and bisexual

21–25

M = 22.82

SD not reported

 

All results control for gender and race/ethnicity

Parent–family rejection

Positively associated with odds of having had unprotected sex with a casual partner in past 6 months

Positively associated with odds of having had unprotected sex at last sex with a casual partner

Not associated with having ever been diagnosed with an STD

Table 2

Parental influences on LGB violence and victimization

Citation

Study design

Sampling strategy

Sample characteristics

Age

Parental influences

Violence/victimization results

D’Augelli et al. (1998)

Cross-sectional

Data analyzed from a larger sample of N = 194 LGB youth; only included youth who lived with parents

Waiver of parental consent obtained

Professional counselor approved by IRB at each site; excluded sites without a trained adult

Self-administered closed-ended survey

Convenience sample

Recruited from LGB social, recreational, and support groups in 14 metropolitan areas in the U.S.

Participation rate and refusal bias not reported

N = 105

71% male

29% female

68% White;

32% youths of color

100% lesbian, gay or bisexual

14–21

M = 18.4

SD = 1.7

Parental knowledge of sexual orientation

Among youth who had disclosed, greater percentage of males than females reported that parents offered to protect them from sexual orientation-related attacks

Lower percentage of family members offered to protect youth from sexual orientation-based attacks among youth who did not disclose compared to youth who disclosed

Lower levels of family verbal abuse and physical threats and attacks among youth who had not disclosed compared to youth who disclosed

D’Augelli et al. (2005a)

Cross-sectional

First wave of longitudinal study of victimization among LGB youth

Data analyzed from larger sample of N = 528 youth

Waiver of parental consent obtained

Youth advocate at each site ensured informed consent

Convenience sample

Recruited from three community-based organizations in New York City and two surrounding suburbs

Participation rate and refusal bias not reported

N = 361

56% male; 44% female

41% Black/African American; 29% Hispanic; 27% White

28% G/L; 20% Bisexual but almost totally G/L; 21% Bisexual but mostly G/L; 15% Bisexual but equally G/L and hetero-sexual; 16% Bisexual but mostly heterosexual; 2 questioning

15–19

M = 17

SD not reported

Parents’ awareness of sexual orientation

Positively correlated with sexual orientation victimization by parents and fear of harassment

Youth whose parents were aware of sexual orientation reported a higher mean level of parental verbal abuse due to sexual orientation than youth whose parents did not know

Youth with aware parents reported a higher mean level of family support and a lower mean level of fear of parental harassment or rejection

D’Augelli et al. (2005b)

Cross-sectional

First phase of a two-year longitudinal study

Subsample of larger study of 528 LGB youth: only included youth living with parents who had raised them

Consent procedures not reported

Face-to-face interview with a master’s level mental health clinician using a structured interview protocol

Convenience sample

Recruited from drop-in centers of 3 community-based organizations providing social, educational and recreational services to LGB youth in New York City and two suburbs

Participation rate and refusal bias not reported

N = 293

47% male; 53% female

43% Hispanic: 85% White; 10% African American; 5% Mixed

57% Non-Hispanic: 41% White; 34% Black/ African American; 5% Asian / Pacific Islander; 19% Mixed; n = 1 American Indian; n = 2 unavailable

23% G/L; 20% Bisexual but almost totally G/L; 21% Bisexual but mostly G/L; 17% Bisexual, but equally G/L and hetero-sexual; 19% Bisexual but mostly heterosexual; 1 questioning

15–19

M = 16.83

SD = 1.21

Male

M = 16.79

SD = 1.25

Female

M = 16.86

SD = 1.17

Parents’ awareness of child’s sexual orientation

No differences in parents’ psychological abuse between youth with aware parents and youth with unaware parents

Youth with aware parents reported a higher mean level of parental victimization due to sexual orientation than youth with unaware parents

Youth with aware parents reported a lower mean level of fear of parental harassment or rejection than youth with unaware parents

D’Augelli et al. (2006)

Cross-sectional

First phase of a two-year longitudinal study on sexual orientation related victimization

Waiver of parental consent obtained

Youth advocate available to answer questions

Face-to-face interview with a same-sex master’s level clinician

Convenience sample

Recruited from 3 community-based organizations: 1 in New York City and 2 in New York City suburbs; and snowball sampling to diversify sample

Participation rate and refusal bias not reported

N = 528

52% male; 48% female

Ethnicity: 45% Hispanic; 55% non-Hispanic

Race: 62% White; 25% Black/African American; 10% Multi-racial; 3% Asian; 2 American Indian/Alaskan Native

Lesbian, gay, and bisexual youth

15–19 

M = 17.03

SD = 1.27

Parents called youth a sissy or a tomboy

Youth whose parents called them a sissy or a tomboy had a higher mean level of lifetime victimization and physical victimization than youth whose parents did not

Youth whose parents called them a sissy or a tomboy had an earlier mean age of first verbal victimization and first physical victimization than youth whose parents did not

Not associated with lifetime sexual victimization or with age of first sexual victimization

Parents discouraged gender atypicality

Youth whose parents discouraged gender atypicality had a higher mean level of lifetime victimization and physical victimization than youth whose parents did not

Youth whose parents discourage gender atypicality had an earlier mean age of first physical victimization than youth whose parents did not

Not associated with lifetime sexual victimization or with age of first verbal or first sexual victimization

D’Augelli et al. (2008)

Cross-sectional

First phase of a two-year longitudinal study

Subsample of larger study of 528 LGB youth: only included youth living with parents who had raised them

Waiver of parental consent obtained

Face-to-face interview with a master’s level mental health clinician using a structured interview protocol

Convenience sample

Recruited from drop-in centers of 3 community-based organizations providing social, educational and recreational services to LGB youth in New York City and two suburbs

Participation rate and refusal bias not reported

N = 516

52% male; 48% female

44% Hispanic: 86% White; 11% African American; 3% Mixed

55% Non-Hispanic: 44% White; 35% Black/ African American; 5% Asian/ Pacific Islander; 19% Mixed, n = 1 American Indian, n = 2 unavailable

28% totally G/L; 20% almost totally G/L; 21% Bisexual but mostly G/L; 21% Bisexual but equally G/L and heterosexual; 16% Bisexual but mostly heterosexual; n = 2 uncertain / questioning

15–19 

M = 17.03

SD = 1.21

Males

M = 16.79

SD = 1.25

Females

M = 16.86

SD = 1.17

Parental knowledge of sexual orientation

Youth with aware parents reported a higher mean level of sexual orientation victimization from parents and from siblings

Youth with aware parents reported a lower mean level of fear of rejection from parents and from siblings

Youth with aware parents reported a higher mean level of family support

Positive parental responses to youth sexual orientation

Youth reporting positive parental responses reported a lower mean level of lifetime sexual orientation victimization from parents than youth with negative parents

Youth reporting positive parental responses reported a higher mean level of family support

Table 3

Parental influences on LGB substance use

Citation

Study design

Sampling strategy

Sample characteristics

Age

Parental influences

Substance use results

Espelage et al. (2008)

Cross-sectional

2000 Dane County Youth Survey: students from 18 high schools in a Midwestern county

Passive parental consent and adolescent assent

Self-administered closed-ended survey given during single class period

“Saturated” sample: county-wide school-based sample of all youth in school on day of survey

Participating schools returned surveys for 90–95% of student population

Refusal bias not reported

N = 13,921

49.7% male

50.3% female

78.6% White; 5.4% Biracial; 4.8% Asian; 4.8% Black; 3.6% Hispanic

7.7% lesbian, gay or bisexual; 6.7% questioning; 86% heterosexual

High school students

M = 15.8

SD not reported

Parental support as a moderator of homophobic teasing

Parental support moderated the relationship between homophobic teasing and alcohol-marijuana use among LGB and questioning students

Students with high levels of homophobic teasing and low levels of parental support reported the highest levels of alcohol-marijuana use

Needham and Austin (2010)

Cross-sectional

Add Health: Subsample of Wave III in-home interviews; excluded young adults if married; missing data on sexual orientation; no contact with parents or parent-like figures at Wave III

Wave III: Young adult consent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Subsample: 73.4% of all youth interviewed at Wave III

Wave III participation rate and refusal bias not reported

N = 11,153

49.4% male; 50.6% female

121 Gay males: 53.8% White; 46.2% Non-White

40 Bisexual males: 55% White; 45% Non-White

5,352 Heterosexual males: 55.3% White; 44.7% Non-White

72 Lesbians: 56% White; 43% Non-White.

152 Bisexual females: 67.1% White; 32.9% Non-White

5,416 Heterosexual females: 54.3% White; 45.7% Non-White

18–26

Males

Gay

M = 22.2

SD = 1.60

Bisexual

M = 21.3

SD = 1.59

Heterosexual

M = 21.8

SD = 1.73

Females

Lesbian

M = 21.8

SD = 1.75

Bisexual

M = 21.5

SD = 1.67

Heterosexual

M = 21.6

SD = 1.73

 

Separate analysis for males and females. All results control for age, race/ethnicity, parental education, and living at home at Wave III

Parental support as a mediator between sexual orientation and substance use

For males, there was no association between sexual orientation and substance use

For lesbians, there was no association between sexual orientation and frequent heavy drinking

Among bisexual females, parental support did not mediate the association between sexual orientation and frequent heavy drinking

Among bisexual and lesbian females, parental support partially mediated the association between sexual orientation and marijuana use

Among bisexual and lesbian females, parental support partially mediated the association between sexual orientation and hard drug use

Padilla et al. (2010)

Cross-sectional

Internet survey by OutProud: The National Coalition for Gay, Lesbian, Bisexual and Transgender Youth

Subsample from larger sample of 6,872 youths aged 25 and under

Consent procedures not reported: no identifying information collected

Self-administered survey completed online

Convenience sample

Survey links available on LGB websites, including OutProud, Youth Action, Out in American, Beautiful Boy, and others

Participation rate and refusal bias not reported

N = 1,906

Gender not reported

81% White; 19% Other

55% gay; 25% lesbian; 34% bisexual

12–17 

M = 16

SD not reported

 

All results control for demographic factors, self-esteem, suicidal ideation, community involvement and other parent factors

Neither parent knows sexual orientation

Not associated with odds of using illegal drugs

Mother’s positive reaction to coming out

Negatively associated with odds of using illegal drugs

Father’s positive reaction to coming out

Not associated with odds of using illegal drugs

Parental religion is a barrier to coming out

Not associated with odds of using illegal drugs

Resnick et al. (1997)

Cross-sectional

Add Health: Wave I in-home interviews; excluded youth reporting sexual debut before age 11

Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

79.5% participation rate; refusal bias not reported

N = 11,572

Gender not reported

Race/ethnicity not reported: nationally representative sample

Sexual orientation not reported

7th–12th grade students

Mean age not reported

 

All results control for same-sex sexual attraction / behaviors, demographic, school, individual, and other parent/family factors:

Parent–family connectedness

Negatively associated with alcohol use and marijuana for all 9th–12th grade youth

Parental presence (waking up, after school, at dinner and bedtime)

Negatively associated with alcohol use and marijuana for all 9th–12th grade youth

Ryan et al. (2009)

Retrospective

Family Acceptance Project

Subsample from larger study (N = 245); excluded transgender young adults (n = 21)

Consent procedures not reported

Computer assisted or pencil-and-paper self-administered surveys

Convenience sample

Recruited from 249 LGB venues located 100 miles of project office in urban city: 50% clubs/bars and 50% social service and community organizations

Participation rate and refusal bias not reported

N = 224

51% male; 49% female

48% non-Latino White; 52% Latino

Lesbian, gay and bisexual

21–25 

M = 22.82

SD not reported

 

All results control for gender and race/ethnicity

Parent–family rejection

Positively associated with odds of having used illicit substances in the past 6 months

Positively associated with odds of having had any substance-related problems

Not associated with the odds of having engaged in heavy drinking in the past 6 months

Table 4

Parental influences on LGB mental health and well-being

Citation

Study design

Sampling strategy

Sample characteristics

Age

Parental influences

Mental health and well-being

D’Augelli (2002)

Cross-sectional

Two separate waves of data collection:

 Wave I: 1987–1989

 Wave II: 1995–1997

Combined two waves of data

Waiver of parental consent obtained

Adults at each group vetted by IRB to assure informed consent compliance

Self-administered closed-ended survey completed in group settings

Convenience sample

Wave I: Sent letters to adult group coordinators of LGB youth serving organizations. Interested groups contacted researchers

Wave II: Posted project description on internet for LGB youth serving organizations. Identified adult contact for each group

Participation rate and refusal bias not reported

N = 542

62% male; 38% female

More than 75% White; 8% African American; 4% Hispanic; 1% Native American or Canadian; remaining % identified as multiple backgrounds

Youth from U.S., Canada and New Zealand

74% gay or lesbian; 20% bisexual, mostly gay or lesbian; 6% bisexual but equally gay/lesbian and heterosexual

14–21 

M = 19.08

SD = 1.5

Parental knowledge of child’s sexual orientation

Youth reporting neither parent knows had higher psychoticism and interpersonal sensitivity scores on Brief Symptom Inventory (BSI) than youth with two knowing parents or one knowing parent

Parental rejection of child’s sexual orientation

No relationship with other BSI subscales or General Severity Index (GSI)

Youth reporting both parents reject had higher scores on GSI and on all BSI subscales than youth with two accepting parents

Youth reporting both parents reject had higher scores on GSI and BSI anxiety, phobic anxiety, and somatization subscales than youth with one accepting and one rejecting parent

No association with GSI, controlling for SOV, fear of losing friends, and fears of physical or verbal abuse at home and school

D’Augelliet al. (2005b)

Cross-sectional

First phase of a two-year longitudinal study

Subsample of larger study of 528 LGB youth: only included youth living with parents who had raised them

Consent procedures not reported

Face-to-face interview with a master’s level mental health clinician using a structured interview protocol

Convenience sample

Recruited from drop-in centers of 3 community-based organizations providing social, educational and recreational services to LGB youth in New York City and two suburbs

Participation rate and refusal bias not reported

N = 293

47% male; 53% female

43% Hispanic: 85% White; 10% African American; 5% Mixed

57% Non-Hispanic: 41% White; 34% African American/Black; 5% Asian/Pacific Islander; 19% Mixed; 1 American Indian; 2 no information

23% G/L; 20% bisexual but almost totally G/L; 21% bisexual but mostly G/L; 17% bisexual, but equally G/L and hetero-sexual; 19% bisexual but mostly heterosexual; 1 questioning

15–19 

M = 16.83

SD = 1.21

Parental awareness of child’s sexual orientation

No difference in BSI scores between youth whose parents were aware and youth whose parents were unaware

No difference in self-esteem between youth whose parents were aware and youth whose parents were unaware

D’Augelli et al. (2006)

Cross-sectional

First phase of a two-year longitudinal study on SOV

Waiver of parental consent obtained

Youth advocate available to answer questions

Face-to-face interview with a same-sex master’s level clinician

Convenience sample

Recruited from 3 community-based organizations: 1 in New York City and 2 in suburbs of New York City

Snowball sampling to diversify sample

Participation rate and refusal bias not reported

N = 528

52% male; 48% female

Ethnicity: 45% Hispanic; 55% non-Hispanic

Race: 62% White; 25% Black/African American; 10% Multi-racial; 3% Asian; n = 2 American Indian/ Alaskan Native

Lesbian, gay and bisexual youth

15–19

M = 17.03

SD = 1.27

Parents called youth a sissy or a tomboy

Youth whose parents called them a sissy or tomboy had a higher mean score on BSI than youth whose parents did not

Youth whose parents called them a sissy or tomboy had a higher mean score on Trauma Symptom Checklist (TSC) than youth whose parents did not

Males whose parents called them a sissy had higher mean level of distress about first experience of sexual SOV than males whose parents did not (small n for sexual SOV)

Females whose parents called them a tomboy had a lower mean level of distress about first experience of sexual SOV than females whose parents did not (small n for sexual SOV)

Not associated with distress about first verbal, physical or sexual SOV

Positively associated with PTSD

     

Parents discouraged gender atypicality

Youth whose parents discouraged gender atypicality had higher mean BSI score than youth whose parents did not

      

Youth whose parents discouraged gender atypicality had higher mean score on TSC than youth whose parents did not

Not associated with distress at first verbal, physical or sexual victimization SOV

D’Augelli and Hershberger (1993)

Cross-sectional

Waiver of parental consent obtained

Adult with professional counseling experience vetted by IRB at each site to ensure informed consent; excluded sites without a trained adult

Self-administered closed-ended survey mailed to groups and

completed in group setting supervised by adult

Convenience sample

Recruited via 14 LGB community centers with informal support groups. Only one center per city, representing major urban areas in U.S.

44% return rate on surveys

Refusal bias not reported

N = 194

73% male; 27% female

66% White; 14% African American; 5% Asian American; 6% Hispanic American; 4% American Indian

75% lesbian or gay; 19% bisexual, but mostly lesbian or gay; 6% bisexual and equally lesbian/gay and heterosexual

15–21 

M = 18.9

SD = 1.6

Males

M = 19.2

SD = 1.6

Females

M = 18.1

SD = 1.5

Parental knowledge and acceptance of child’s sexual orientation

Not associated with BSI

Espelage et al. (2008)

Cross-sectional

2000 Dane County Youth Survey: students from 18 high schools in a Midwestern county

Passive parental consent and adolescent assent

Self-administered closed-ended survey given during single class period

“Saturated” sample: county-wide school-based sample of all youth in school on day of survey

Participating schools returned surveys for 90–95% of student population

Refusal bias not reported

N = 13,921

49.7% male

50.3% female

78.6% White; 5.4% Biracial; 4.8% Asian; 4.8% Black; 3.6% Hispanic

7.7% lesbian, gay or bisexual; 6.7% questioning; 86% heterosexual

High school students

M = 15.8

SD not reported

Parental support as a moderator of homophobic teasing

Did not moderate the association between homophobic teasing and depression-suicidal feelings among LGB and questioning students

Floyd et al. (1999)

Cross-sectional

Waiver of parental consent obtained

Two part interviews: Part 1 consisted of structured interviews with open- and closed-ended questions. Part 2 consisted of a self-administered closed-ended survey

Convenience sample

Recruited from youth support groups, flyers, newspaper ads and articles, and campus organizations in the upper Midwest and Southeast

Participation rate and refusal bias not reported

N = 72

50% male; 50% female

79% White; 7% Asian-American; 6% African-American; 3% Native American; 6% other

Lesbian, gay or bisexual

16–27 

M = 20.88

SD = 2.94

Mother’s positive attitudes about child’s sexual orientation

Positively correlated with self-esteem

Negatively correlated with BSI depression

Negatively correlated with BSI anxiety for daughters

Not correlated with BSI hostility

Father’s positive attitudes about child’s sexual orientation

Not correlated with self-esteem, BSI depression, anxiety, and hostility

     

Relatedness to mother

Positively correlated with self-esteem

      

Not correlated with BSI hostility

      

Negatively correlated with BSI anxiety and depression

     

Relatedness to father

Positively correlated with self-esteem

      

Negatively correlated with BSI anxiety, depression, and hostility

     

Autonomy from mother

Not correlated with self-esteem, BSI anxiety, depression, and hostility

     

Autonomy from mother moderated by relatedness to mother

Negatively correlated with self-esteem only when maternal relatedness was low

     

Autonomy from mother moderated by youth gender

Positively correlated with self-esteem for daughters but not sons

     

Autonomy from father

Positively correlated with self-esteem

      

Not correlated with BSI anxiety, depression, and hostility

     

Less conflictual independence with mother

Positively correlated with self-esteem

Negatively correlated with BSI depression

Not correlated with BSI anxiety or hostility

     

Less conflictual independence with mother moderated by youth gender

Negatively correlated with BSI hostility for sons but not daughters

     

Less conflictual independence with father

Negatively correlated with BSI anxiety and hostility

      

Not correlated with self-esteem or BSI depression

     

Less conflictual independence with father moderated by youth age

Negatively correlated with BSI hostility for younger adolescents but not for older adolescents

Homma and Saewyc (2007)

Cross-sectional

2001 MN Student Survey

Subsample of 9th and 12th grade Asian American students with a same-sex sex partner in past year

Consent procedures not reported

Paper-and-pencil survey completed in classroom

“Saturated” sample: state-wide school-based sample of all youth in school on day of survey

Survey of all public school students in grades 6, 9, and 12

92% of schools participated: represents 97% of MN students in grades 6, 9, and 12

Refusal bias not reported

N = 91

63% male; 37% female

84% of males reported both-gender partners

76% of females reported both-gender partners

13–19

Mean age not reported

56% 9th graders

44% 12th graders

Parent–family caring

Low levels of parental caring were associated with low levels of self-esteem, which were associated with greater emotional distress (e.g., self-esteem mediated the association between perceived parental caring and emotional distress)

Maguen et al. (2002)

Cross-sectional

Waiver of parental consent obtained. Adult at each site asked to serve in loco parentis

Adolescents were not asked to provide name for consent

Self-administered closed-ended survey completed at either conference or community center

Convenience

Multiple strategies: recruited from LGBT conference at a southeastern university (n = 103) and from southeastern GL community service center (n = 14)

98% participation rate

Refusal bias not reported

N = 117

54% male; 46% female

75% European American; 11% African American; 5% Latino; 4% Biracial; 3% Asian; and 1% Other; n = 1 unreported

46% gay; 29% lesbian; 22% bisexual; 3% queer; n = 1 unreported

14–27 

M = 20

SD = 2.78

Parental knowledge of child’s sexual orientation

No differences in mean self-esteem if youth reported that one parent knew or that two parents knew their sexual orientation

Youth who reported that their fathers knew their sexual orientation had a higher mean level of self-esteem than youth whose fathers did not know

Needham and Austin (2010)

Cross-sectional

Add Health: Subsample of Wave III in-home interviews; excluded young adults if married; missing data on sexual orientation; no contact with parents or parent-like figures at Wave III

Wave III: Young adult consent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Subsample: 73.4% of all youth interviewed at Wave III

Wave III participation rate and refusal bias not reported

N = 11,153

49.4% male; 50.6% female

121 Gay males: 53.8% White; 46.2% Non-White

40 Bisexual males: 55% White; 45% Non-White

5,352 Heterosexual males: 55.3% White; 44.7% Non-White

72 Lesbians: 56% White; 43% Non-White.

152 Bisexual females: 67.1% White; 32.9% Non-White

5,416 Heterosexual females: 54.3% White; 45.7% Non-White

18–26

Males

Gay

M = 22.2

SD = 1.60

Bisexual

M = 21.3

SD = 1.59

Heterosexual

M = 21.8

SD = 1.73

Females

Lesbian

M = 21.8

SD = 1.75

Bisexual

M = 21.5

SD = 1.67

Heterosexual

M = 21.6

SD = 1.73

 

All results control for age, race/ ethnicity, parental education, and living in parental home at Wave III

Parental support

Association between sexual orientation and depression was fully mediated by parental support for bisexual but not lesbian females

Among males in the subsample, there were no mean differences in depression at Wave III as a function of sexual orientation

Resnick et al. (1997)

Cross-sectional

Add Health: Wave I in-home interviews; excluded youth reporting sexual debut before age 11

Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

79.5% participation rate; refusal bias not reported

N = 11,572

Gender not reported

Race/ethnicity not reported: nationally representative sample

Sexual orientation not reported

7th–12th grade students

Mean age not reported

 

All results control for same-sex sexual attraction/behaviors, demographic factors, school factors, individual factors, and other parent and family factors

Parent–family connectedness

Negatively associated with emotional distress for all 9th–12th grade youth

Parent–adolescent shared activities

Positively associated with emotional distress for all 9th–12th grade youth

Parental presence (waking up, after school, at dinner and bedtime)

Negatively associated with emotional distress for all 9th–12th grade youth

Parental school expectations

Negatively associated with emotional distress for all 9th–12th grade youth

Ryan et al. (2009)

Retrospective

Family Acceptance Project

Subsample from larger study (N = 245); excluded transgender young adults (n = 21)

Consent procedures not reported

Computer assisted or pencil-and-paper self-administered surveys

Convenience sample

Recruited from 249 LGB venues located 100 miles of project office in urban city: 50% clubs/bars and 50% social service and community organizations

Participation rate and refusal bias not reported

N = 224 lesbian, gay and bisexual young adults

51% male;

49% female

48% non-Latino White; 52% Latino

21–25 

M = 22.82

SD not reported

 

Results control for gender and race/ethnicity

Parent–family rejection

Positively associated with odds of being currently depressed

Savin-Williams (1989a)

Cross-sectional

Self-administered closed-ended survey

Convenience sample

Multiple recruitment strategies: LGB picnic; LGB college campus meetings; GL activist conference; college course; and snowball sampling in local and out-of-state friendship networks

Participation rates:

LGB Picnic: Not reported; 7% of sample

Campus meetings: 100% participation rate; 15% of sample

Conference: 92.5% participation rate; 20% of sample

College course: Not reported; 2% of sample

Snowball sampling: not reported; Local: 43% of sample; Out-of-state: 14% of sample

Refusal bias not reported

N = 317

68% male; 32% female

91% White; 9% Not reported

68% gay males; 32% lesbian

14–23

Mean age not reported

 

Separate analyses for male versus female youth and parent gender. Results control for other parental influences

Maternal knowledge of sexual orientation

Not associated with self-esteem for females

Positively associated with self-esteem for rural males

Satisfaction with maternal relationship

Positively associated with self-esteem for all youth (males and females)

Infrequent contact with mother

Not associated with self-esteem for all females

Positively associated with self-esteem for urban males

Paternal knowledge of sexual orientation

Not associated with self-esteem for all youth (males and females)

Satisfaction with paternal relationship

Not associated with self-esteem for all females

Positively associated with self-esteem for all males

Infrequent contact with father

Not associated with self-esteem for all females

Positively associated with self-esteem for all males

Savin-Williams (1989b)

Cross-sectional

Self-administered closed-ended survey

Multiple recruitment strategies: LGB picnic; LGB college campus meetings; GL activist conference; college course; and snowball sampling in local and out-of-state friendship networks

Participation rates:

LGB Picnic: Not reported; 7% of sample

Campus meetings: 100% participation rate; 15% of sample

Conference: 92.5% participation rate; 20% of sample

College course: Not reported; 2% of sample

Snowball sampling: not reported; Local: 43% of sample; Out-of-state: 14% of sample

Refusal bias not reported

N = 317 adolescents and young adults

68% male; 32% female

91% White; 3% Hispanic; 3% Black; 2% Asian American; 1% Native American; 1% International

46% exclusively homosexual; 31% predominantly homosexual; 33% reported both homosexual and heterosexual interests

14–23

Mean age not reported

3% high school: 14–17 years old

70% college: 17–23 years old

8% in grad school: 21–23 years old

19% not in school: 16–23 years old

 

Separate analyses for youth and parent gender

Maternal acceptance of sexual orientation moderated by youth comfort with being gay

Not associated with self-esteem for females

Maternal acceptance moderated by perceived maternal importance for youth self-worth

Not associated with self-esteem for females

Paternal acceptance moderated by youth comfort

Not associated with self-esteem for females

Paternal acceptance moderated by perceived paternal importance

Positively associated with self-esteem for females

Maternal acceptance mediated by perceived maternal importance

Direct effect of acceptance on the self-esteem of males is completely mediated by perceived maternal importance

Paternal acceptance mediated by perceived paternal importance

Direct effect of acceptance on the self-esteem of males is completely mediated by perceived paternal importance

Sheets and Mohr (2009)

Cross-sectional

Self-administered internet survey

Subsample from larger survey of 301 young adults: excluded if missing data, outside age range, not bisexual, transgender, or failed survey validity checks

Convenience sample

Recruited from electronic mailing lists of LGBT student organizations on 32 public university/ college campuses representing all major regions of U.S.

Electronic signature of consent

N = 210 young adults

85% female; 15% male

81% White; 5% Black / African American; 4% Hispanic; 3% Asian or Pacific Islander; 1% Native American; 6% Other

18–25 

M = 20.96

SD = 1.77

 

Results control for gender and other social support factors

Parent–family general social support (GSS)

Negatively associated with depression

Parent–family sexuality specific support (SSS)

Not associated with depression

Parent–family GSS as a moderator of friend GSS

Did not moderate relationship between friend GSS and depression

Parent–family SSS as a moderator of friend SSS

Did not moderate relationship between friend SSS and depression

Teasdale and Bradley-Engen (2010)

Prospective

Add Health: Wave I and II in-home interviews

Subsample of same-sex attracted (SSA) and heterosexual youth

Waves I and II: Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Participation rate and refusal bias not reported

N = 11,911

48% male; 52% female

42% White, 20% Black, 24% Hispanic, 14% Other

7% Same-Sex Attracted (n = 787)

98% Heterosexual (n = 101,124)

Grades 7–12 

M = 15.90

SD = 1.54

 

Results control for demographic factors and for peer and school support variables

Perceived parental caring

Negatively associated with depression for SSA youth

Partially mediated relationship between sexual orientation and depression for SSA youth

Mediated relationship between stress at home and depression for SSA youth

Ueno (2005)

Prospective

Add Health: Wave I and II in-home interviews

Subsample of youth who were older than age 13 and classified as sexual minority youth based on reports of same-sex or both-sex attraction or dating; or sexual majority youth based on reports of only opposite-sex attraction or dating. Excluded youth reporting no attraction or dating experience

Waves I and II: Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Participation rate and refusal bias not reported

N = 11,571

Sexual Minority Youth

7.67% (n = 887)

56% male; 44% female

63% White; 18% Black; 13% Hispanic; 3% Asian; 3% Other Sexual Majority Youth

92.33% (n = 10,684)

56% male; 44% female

69% White; 15% Black; 12% Hispanic; 3% Asian; 1% Other.

Grades 7–12

Sexual minority:

M = 15.81

SD = 1.56

Sexual majority:

M = 15.61

SD = 1.61

Arguing with parents

Positively associated with psychological distress for all youth

Attachment to parents

Negatively associated with psychological distress for all youth

Arguing with parents moderated by number of friends moderated by sexual minority status

No statistically significant difference in psychological distress between sexual minority and sexual majority youth

Arguing with parents moderated by number of sexual minority friends moderated by sexual minority status

Number of sexual minority friends buffers against the negative effect of arguing with parents on psychological distress for sexual minority but not for sexual majority youth

Table 5

Parental influences on LGB suicide

Citation

Study design

Sampling strategy

Sample characteristics

Age

Parental influences

Suicide results

D’Augelli et al. (1998)

Cross-sectional

Data analyzed from larger sample of N = 194 LGB youth; only included youth who lived with parents

Waiver of parental consent obtained

Professional counselor approved by IRB at each site; excluded sites without a trained adult

Self-administered closed-ended survey

Convenience sample

Recruited from LGB social, recreational, and support groups in 14 metropolitan areas in the U.S.

Participation rate and refusal bias not reported

N = 105

71% male

29% female

68% White;

32% youths of color

100% lesbian, gay or bisexual

14–21 

M = 18.4

SD = 1.7

Parental knowledge of child’s sexual orientation

Higher percentage of suicide attempts among youth who had disclosed sexual orientation to parents compared to youth who had not disclosed

Greater frequency of suicidal thoughts among youth who had disclosed their sexual orientation to their parents compared to youth who had not disclosed

D’Augelli et al. (2005a)

Cross-sectional

First wave of longitudinal study of victimization among LGB youth

Data analyzed from larger sample of N = 528 youth.

Waiver of parental consent obtained

Youth advocate at each site ensured informed consent

Convenience Sample

Recruited from three community-based organizations in New York City and two surrounding suburbs

Participation rate and refusal bias not reported

N = 361

56% male; 44% female

41% Black/African American; 29% Hispanic; 27% White

28% G/L; 20% Bisexual but almost totally G/L; 21% Bisexual but mostly G/L; 15% Bisexual but equally G/L and heterosexual; 16% Bisexual but mostly heterosexual; 2 questioning

15–19 

M = 17

SD = not reported

 

Results control for demographic factors, sexual orientation factors, and other parent and family factors

Parent’s psychological abuse

Discriminated sexual orientation-related (SOR) suicide attempters from non-SOR suicide attempters and non-attempters

Parents discouraged gender-atypical behavior

Discriminated SOR suicide attempters from non-SOR suicide attempters and non-attempters

Parent calls child LGB

Did not discriminate SOR suicide attempters from non-SOR suicide attempters and non-attempters

Parents called youth a sissy or a tomboy

Discriminated SOR suicide attempters from non-SOR suicide attempters and non-attempters

D’Augelli and Hershberger (1993)

Cross-sectional

Waiver of parental consent obtained

Adult with professional counseling experience vetted by IRB at each site to ensure informed consent; excluded sites without a trained adult

Self-administered closed-ended survey mailed to groups and completed in group setting supervised by adult

Convenience sample

Recruited via 14 LGB community centers with informal support groups. Only one center per city, representing major urban areas in U.S.

44% return rate on surveys

Refusal bias not reported

N = 194

73% male; 27% female

66% White; 14% African American; 5% Asian American; 6% Hispanic American; 4% American Indian

75% lesbian or gay; 19% bisexual, but mostly lesbian or gay; 6% bisexual and equally lesbian/gay and heterosexual

15–21 

M = 18.9

SD = 1.6

Males

M = 19.2

SD = 1.6

Females

M = 18.1

SD = 1.5

Parental awareness of child’s sexual orientation

Lack of parental awareness positively associated with suicidal thoughts

Compared to suicide non-attempters, suicide attempters were more likely to have disclosed sexual orientation to a family member other than a parent

Compared to suicide attempters, parents of non-attempters were less aware of their child’s sexual orientation

Parents’ reaction (positive or negative) to disclosure of sexual orientation

Did not differentiate between suicide attempters and non-attempters

D’Augelli et al. (2001)

Cross-sectional

Data from project examining how LGB youth cope with sexual orientation-related challenges. Excluded data from mostly heterosexual, hetero-sexual, uncertain, no answer youth

IRB approved adult required to be at each site to ensure informed consent. Parental consent procedures not reported

Self-administered closed-ended survey completed in group setting

Convenience sample

Recruited from 39 community-based organizations and 20 colleges in U.S., Canada, and New Zealand using multiple recruitment strategies: letters, internet postings, and telephone

Participation rate and refusal bias not reported

N = 350

56% male; 44% female

78% White; 8% African American; 4% Asian; 3% Chicano or Mexican; 7% Other

Males:

83% gay; 17% bisexual

Females:

64% lesbian; 36% bisexual

14–21

Mean age not reported

Parental knowledge of child’s sexual orientation

Among youth reporting suicide attempts, 54% occurred before a parent knew sexual orientation, 26% occurred during same year as disclosure to a parent, and 20% occurred within a year of disclosure to a parent

Parental rejection or intolerance of youth’s sexual orientation

Youth with intolerant or rejecting fathers were twice as likely to report a past suicide attempt

Of suicide attempters, 48% reported intolerant or rejecting fathers, compared to 28% of non-suicide attempting youth

No difference in maternal rejection or intolerance between suicide attempters and non-suicide attempters

Espelage et al. (2008)

Cross-sectional

2000 Dane County Youth Survey: students from 18 high schools in a Midwestern county

Passive parental consent and adolescent assent

Self-administered closed-ended survey given during single class period

“Saturated” sample: county-wide school-based sample of all youth in school on day of survey

Participating schools returned surveys for 90–95% of student population

Refusal bias not reported

N = 13,921

49.7% male; 50.3% female

78.6% White; 5.4% Biracial; 4.8% Asian; 4.8% Black; 3.6% Hispanic

7.7% lesbian, gay or bi; 6.7% questioning; 86% heterosexual

High school students

M = 15.8

SD not reported

Homophobic teasing moderated by parental support

Parental support did moderate the relationships between homophobic teasing and depression-suicidal feelings among LGB and questioning students

Eisenberg and Resnick (2006)

Cross-sectional

2004 Minnesota Student Survey

Subsample of 6th, 9th and 12th grade students; excluded youth reporting no sexual partners

Passive parental consent (active if required by school or district)

Self-administered closed-ended survey during class period

“Saturated” sample: state-wide school-based sample of all youth in school on day of survey

88% of school districts in Minnesota participated

Statewide: 75% of 9th grade and 55% of 12 grade students participated

Refusal bias not reported

N = 21,927

10.28% LGB (n = 2,255)

64% male; 36% female

67% White; 6% Black / African American; 5% Hispanic; 5% Asian; 2% Native American; 15% Mixed/Other

90.13% heterosexual

(n = 19,672)

47% male; 53% female

81 White; 4% Black / African American; 3% Hispanic; 3% Asian; 1% Native American; 7% Mixed/Other

Grades 9 and 12

Full sample:

35.3% 9th grade

65.7% 12th grade

 

Results control for sexual orientation, demographic factors, teacher caring, other adult caring, and school safety

Parent–family connectedness

Parent–family connectedness negatively associated with the odds of suicidal ideation and suicidal attempts

Parent–family connectedness accounted for a greater amount of variance in suicidal behaviors than sexual orientation or any other protective factor

Friedman et al. (2006)

Retrospective

Consent procedures not reported

Self-administered closed-ended survey

Convenience sample

Recruited from gay university- or community-based organizations

Participation rate and refusal bias not reported

N = 96

100% male

73% White; 10% African-American; 6% Latino; 5% Asian / Pacific Islander; 6% Mixed

88% gay; 8% bisexual; 4% other

18–25 

M = 20.32

SD = 1.83

Parental social support

Higher levels of parental support in elementary, junior and high school independently protected against current suicidality

Parental social support as a moderator of bullying and suicidality

Parental social support did not moderate the relationship between bullying and suicidality

Hershberger et al. (1997)

Cross-sectional

Adult human services professional at each site explained study, ensured informed consent and administered survey

Self-administered closed-ended survey

Convenience sample

Recruited from 14 youth groups in LG community centers in major U.S. urban areas

44% survey return rate

Human service professional at each site reported that no youth given a survey refused participation

N = 104

73% male; 27% female

66% White; 15% African American; 5% Asian American; 5% Hispanic; 4% Native American

15–21 

M = 18.86

SD = 1.64

Males

M = 19.16

SD = 1.57

Females

M = 18.06

SD = 1.54

Mother and father knows sexual orientation

Associated with a higher odds of having attempting suicide

Not associated with suicide attempts, controlling for demographic factors, sexual orientation experiences and behaviors, victimization, disclosure, and mental health

Mother knows sexual orientation

Did not distinguish between non-attempters and single and multiple suicide attempters

Father knows sexual orientation

Distinguished between non-attempters and single and multiple suicide attempters

Needham and Austin (2010)

Cross-sectional

Add Health: Subsample of Wave III in-home interviews; excluded young adults if married; missing data on sexual orientation; no contact with parents or parent-like figures at Wave III

Wave III: Young adult consent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Subsample: 73.4% of all youth interviewed at Wave III

Wave III participation rate and refusal bias not reported

N = 11,153 49.4% male; 50.6% female Gay males (n = 121)

53.8% White; 46.2% Non-White

Bisexual males (n = 40): 55% White; 45% Non-White

Heterosexual males (n = 5,352): 55.3% White; 44.7% Non-White

Lesbians (n = 72): 56% White; 43% Non-White.

Bisexual females (n = 152): 67.1% White; 32.9% Non-White

Heterosexual females (n = 5,416): 54.3% White; 45.7% Non-White

18–26

Males

Gay

M = 22.2

SD = 1.60

Bisexual

M = 21.3

SD = 1.59

Heterosexual

M = 21.8

SD = 1.73

Females

Lesbian

M = 21.8

SD = 1.75

Bisexual

M = 21.5

SD = 1.67

Heterosexual

M = 21.6

SD = 1.73

 

Separate analyses by gender

All results control for age, race/ ethnicity, parental education, and living in parental home at Wave III

Parental support as a mediator between sexual orientation and suicidal thoughts

Association between sexual orientation and suicidal thoughts was partially mediated by parental support for bisexual and lesbian females

Association between sexual orientation and suicidal thoughts was partially mediated by parental support for gay but not bisexual males

Proctor and Groze (1994)

Cross-sectional

Adult at each site obtained informed consent. Parental consent not reported

Self-administered closed-ended survey completed in youth group setting

Convenience

Recruited via 56 LGB youth groups in US and Canada: groups mailed letter, instrument copy and consent forms

24 groups agreed to participate. No other participation rate reported

Refusal bias not reported

N = 221

71.9% male; 28.1% female

69.1% White; 7.3% African American; 6.8% Latino; 5% Native American; 4.5% Asian; 0.5% Pacific Islander; 6.8% Other

62.9% gay; 23.5% lesbian; 13.6% bisexual

90.5% from U.S.; 9.5% from Canada

Range not reported; No youth over age 21 

M = 18.5

SD not reported

Good parent–child relationship

Having a good relationship discriminated between youth who had neither considered nor attempted suicide and youth who had considered and attempted suicide

Remafedi et al. (1991)

Cross-sectional

Verbal and written consent obtained

Self-administered structured interview

Convenience sample

Multiple recruitment strategies used: ads in gay publications (30% of total sample); bars (5%); LGB youth social support groups (29%); universities (15%); a youth drop-in center (19%); peer referrals (11%)

No recruitment or referrals from mental health treatment facilities

Participation rate and refusal bias not reported

N = 137

100% male

82% White; 13% African American; 4% Hispanic; 1% Asian

88% gay; 12% bisexual

14–21

Suicide attempters

M = 19.25

SD = 1.63

Non-attempters

M = 19.83

SD = 1.63

Maternal knowledge of sexual orientation

Did not discriminate between suicide attempters and non-attempters

Paternal knowledge of sexual orientation

Did not discriminate between suicide attempters and non-attempters

Supportive maternal response

Did not discriminate between suicide attempters and non-attempters

Supportive paternal response

Did not discriminate between suicide attempters and non-attempters

Rotheram-Borus et al. (1994)

Cross-sectional

Lesbian youth excluded due to funding concerns

Voluntary informed consent. Parental consent not reported

Semi-structured face-to-face interview and self-administered closed-ended survey

Convenience sample

Recruited from Hetrick-Martin Institute in New York City

Three youths refused to participate. No refusal bias reported

N = 131

100% male

51% Hispanic; 30% Black; 12% White; 7% Other

66% gay; 25% bisexual; 3% straight; 6% did not identify as gay, bisexual or straight

14–19 

M = 16.8

SD = 1.4

Parental knowledge of sexual orientation

Disclosure of sexual orientation was positively associated with the odds of reporting a past suicide attempt

Parents discovered adolescent was gay

Parental discovery of sexual orientation was positively associated with the odds of reporting a past suicide attempt

Ryan et al. (2009)

Retrospective

Family Acceptance Project

Subsample from larger study (N = 245); excluded transgender young adults (n = 21)

Consent procedures not reported

Computer assisted or pencil-and-paper self-administered surveys

Convenience sample

Recruited from 249 LGB venues located 100 miles of project office in urban city: 50% LGB-serving clubs and bars and 50% LGB social service and community organizations

Participation rate and refusal bias not reported

N = 224 young adults

51% male;

49% female

48% non-Latino White; 52% Latino

21–25 

M = 22.82

SD not reported

 

All results control for gender and race/ethnicity

Parent–family rejection

Associated with a higher odds of suicidal ideation

Associated with a higher likelihood of previous suicide attempts

Teasdale and Bradley-Engen (2010)

Prospective

Add Health: Wave I and II in-home interviews

Subsample of same-sex attracted (SSA) and heterosexual youth

Waves I and II: Parental consent and adolescent assent

CAPI and A-CASI surveys

Multistage, stratified, probability sample

Nationally representative school-based sample

Participation rate and refusal bias not reported

N = 11,911

48% male; 52% female

42% White; 20% Black; 24% Hispanic; 14% Other

7% Same-Sex Attracted (n = 787)

98% Heterosexual (n = 101,124)

Grades 7–12 

M = 15.90

SD = 1.54

 

Results control for demographic factors and for peer and school support variables

Perceived parental caring

Negatively associated with odds of reporting suicidal tendencies for SSA youth

For SSA youth, partially mediated the relationship between sexual orientation and the odds of reporting suicidal tendencies

The total number of articles that examined each health topic was as follows: 6 examined sexual behavior, 5 studied violence and victimization, 5 focused on substance use, 16 examined mental health, and 15 examined suicide. Below, we discuss the results within each targeted health area. Key results are presented in Tables 1, 2, 3, 4, 5. Given the scope of the results, it is not possible to discuss each finding in detail. Within each section, we therefore highlight key trends and limitations within a given health area as well as the types of parental influences that were studied. Results from methodologically rigorous studies are highlighted, as are patterns of results across numerous studies.

Sexual Risk Behaviors, STIs, and HIV

In total, six articles examined parental influences on the sexual risk behaviors and related health outcomes of LGB youth. Table 1 summarizes the key findings. Across the six articles, parental influences were explored in three primary areas: (a) emotional dimensions of the parent–child relationship, namely parent–child connectedness; (b) parental values that discourage risk taking, such as disapproval; and (c) parents’ knowledge of and responses to their child’s sexual orientation. Across these studies, there were several trends. First, only one article reported prospective data (i.e., Ford et al. 2005). Half of the articles reported findings based on probability samples of LGB young people (Ford et al. 2005; O’Donnell et al. 2002; Resnick et al. 1997), whereas the other half relied on convenience samples. In addition, the study by O’Donnell et al. (2002) was the only non-Add Health article in the review to utilize a probability sample of LGB youth. With two nationally representative studies and one statewide survey (i.e., Ackard et al. 2008), the articles included data from urban, rural, and suburban youth.

Of the six articles, two emerged as methodologically superior in that they were based on Add Health data and controlled for a number of potentially important variables (Ford et al. 2005; Resnick et al. 1997). In general, there was modest support for parent–child connectedness and strong support for parental disapproval on LGB youths’ sexual behaviors and health. A protective association for parent–child connectedness was observed by a number of studies (i.e., Ackard et al. 2008; Garofalo et al. 2008; Resnick et al. 1997). More specifically, Resnick et al. (1997) found that connectedness and parental disapproval were associated with an older age of sexual debut when controlling for other protective and demographic factors, including sexual orientation (Resnick et al. 1997). However, in a subsequent longitudinal study replicating Resnick et al.’s (1997) study on the likelihood of being diagnosed with an STI in early adulthood, only parental disapproval was significant (Ford et al. 2005).

In the only other study using a probability sample, there was no support for an association between parental knowledge and the sexual behavior of urban Latino men who have sex with men (MSM; O’Donnell et al. 2002). Although limited by a retrospective study design and the use of an urban convenience sample, Ryan et al. (2009) found that perceived parental/caregiver rejection during adolescence was positively associated with sexual risk behaviors in early adulthood, with high levels of rejection emerging as particularly important.

Substance Use

A total of five articles focused on the influence of parents on substance use among LGB youth. The results are summarized in Table 2 and highlight four trends. First, the included articles focused on two areas of parental influences. These were emotional dimensions of the parent–child relationship, such as support and connectedness (Espelage et al. 2008; Needham and Austin 2010; Resnick et al. 1997), and parental knowledge of and responses to a child’s sexual orientation (Padilla et al. 2010; Ryan et al. 2009). Second, there were no prospective studies, and third, only two articles were based on probability samples (Needham and Austin 2010; Resnick et al. 1997). Despite these limitations, two mechanisms of influence emerged as important across the five studies. These were the benefits of a good parent–child relationship characterized by connectedness, warmth, and support, and the harm associated with parental rejection of youths’ sexual orientation.

Two methodologically strong articles analyzed Add Health data to examine parental influences in adolescence and early adulthood (Needham and Austin 2010; Resnick et al. 1997). Consistent with the idea that parents can buffer their child from sexual orientation-related stressors, Needham and Austin (2010) examined parental support in early adulthood as a potential mediator of the association between sexual orientation and substance use during this same developmental period. Although LGB young adults reported lower mean levels of parental support than their heterosexual peers, parental support partially mediated the association between sexual orientation and substance use among lesbian and bisexual females, with some exceptions (Needham and Austin 2010). Notably, there was no association between sexual orientation and substance use among males (Needham and Austin 2010).

In contrast, parent’s negative responses to their child’s sexual orientation emerged as having a positive association with young adult’s substance use, an association that was particularly strong when young adults reported high levels of parental rejection during adolescence (Ryan et al. 2009). With no prospective studies, it is difficult to draw firm conclusions about the influence of parents on the substance use of LGB youth. While the findings suggest that support, connectedness, and rejection are important correlates of substance use in both adolescence and early adulthood, additional research is needed to better understand the temporal nature of the observed associations.

Violence and Victimization

Five articles were located for parental influences on LGB youths’ experiences with violence and victimization (see Table 3). All five examined how parents’ knowledge of and responses to their child’s sexual orientation were associated with sexual orientation-related victimization (SOV). Considered together, there was some support that these two dimensions of parenting were positively associated with young people’s reports of SOV both within (D’Augelli et al. 1998, 2005a, b) and outside of (D’Augelli et al. 2006) the family context. Notably, only one article examined the role of positive parental responses (D’Augelli et al. 2008), which were negatively associated with SOV and positively associated with youths’ reports of family support. In addition, family support and offers to protect youth from SOV were higher when parents were aware of their child’s sexual orientation (D’Augelli et al. 2005a, 2008).

Taken together, the studies suggested that parents can serve as victimizers and protectors of their children. However, methodological limitations temper the strength of the findings. All of the studies were cross-sectional and utilized convenience samples recruited from LGB-serving organizations located in and around urban, metropolitan areas. Four were based on the same sample of LGB youth (D’Augelli et al. 2005a, b, 2006, 2008) whereas the fifth drew upon LGB youth residing in 14 major metropolitan areas in the U.S. (D’Augelli et al. 1998). None reported participation rates or refusal bias. As such, it is difficult to make causal inferences about the nature of the observed associations or to generalize the findings to youth residing in non-urban settings and to those not attending LGB-serving organizations.

Mental Health and Well-Being

Sixteen articles examined parental influences on the mental health and well-being of LGB youth. Similar to previous areas, the majority focused on two types of parental influences: (a) parents’ knowledge of and responses to their child’s sexual orientation and (b) emotional dimensions of the parent–child relationship. In addition, a very small number examined parental victimization of LGB youth (D’Augelli 2002) and patterns of autonomy and independence in the parent–child relationship (Floyd et al. 1999). Table 4 summarizes the results.

Multiple studies examined how emotional dimensions of the parent–child relationship, such as support, caring, connectedness, and conflict, shaped youths’ mental health. Across studies, a supportive and caring parent–child relationship emerged as an important correlate (Floyd et al. 1999; Homma and Saewyc 2007; Needham and Austin 2010; Resnick et al. 1997; Savin-Williams 1989a, b; Sheets and Mohr 2009; Teasdale and Bradley-Engen 2010; Ueno 2005). The strongest support was offered by Ueno (2005) and Teasdale and Bradley-Engen (2010), who prospectively examined parental influences on youth well-being in Add Health. Teasdale and Bradley-Engen (2010) found that parental support in adolescence was negatively correlated with depression 1 year later and could mediate the relationship between sexual orientation and mental health. Similarly, Ueno (2005) found that parent–child attachment was negatively correlated with youths’ psychological distress whereas arguing with parents was positively correlated. Although LGB youth reported a higher mean level of distress than heterosexual youth, the difference was small in magnitude, leading Ueno (2005) to note that LGB youth may experience less stressful environments than what is portrayed in the literature.

Across separate articles, negative parental responses were inversely associated with young people’s mental well-being (D’Augelli 2002; D’Augelli et al. 2006; Ryan et al. 2009; Savin-Williams 1989b). Mental health was one of the few health areas where studies reported on both mother and father data, with different associations emerging for each parent (Floyd et al. 1999; Savin-Williams 1989a). In addition, data on youths’ perceptions of both parents allowed for an examination of how patterns of parental responses shaped young people’s mental health. Findings indicated that parents often responded differently and that having two accepting parents was associated with better mental health outcomes than having one accepting and one rejecting parent (D’Augelli 2002).

Although the overall pattern of results suggested that negative parental responses and a strong parent–child relationship are both important, there were important limitations. Of the 16 articles, the majority relied on convenience samples (n = 12) and presented cross-sectional data (n = 13). As with previous health areas, probability samples of LGB youth were only found in studies using Add Health data (e.g., Needham and Austin 2010; Resnick et al. 1997; Teasdale and Bradley-Engen 2010; Ueno 2005). Apart from these four studies, both rural and ethnic minority youth were underrepresented, as were youth not attending LGB organizations or social venues.

Suicide

A total of 14 articles examined how parents influence LGB youth’s experiences with suicide. Across the included studies, parental influences were again examined in two primary domains: (a) parental knowledge of and responses to their child’s sexual orientation and (b) emotional dimensions of the parent–child relationship. In addition, one study examined abuse within the parent–child relationship (D’Augelli et al. 2005a). Table 5 presents the results.

A small number of studies examined parents’ negative responses to their children’s sexual orientation, which were positively associated with suicide (D’Augelli et al. 2001; Ryan et al. 2009). A larger number (n = 6) examined the role of parental knowledge of a child’s sexual orientation and tended to indicate that knowledge was positively associated with suicidal thoughts and attempts (D’Augelli and Hershberger 1993; D’Augelli et al. 1998, 2001; Rotheram-Borus et al. 1994). However, the cross-sectional study designs of these studies make it difficult to draw firm conclusions as it is possible that parents became aware of their child’s sexual orientation after their child had expressed suicidal thoughts or attempted suicide. Some support for this alternative was offered by D’Augelli et al. (2001), who found that among LGB youth reporting suicide attempts, 54% of attempts occurred when parents were unaware of their child’s sexual orientation, 26% occurred in the same year as disclosure to a parent (before or after disclosure is unknown), and 20% occurred within a year of disclosing one’s sexual orientation to a parent. The period prior to and immediately following youths’ disclosure of their sexual orientation can be a stressful time for LGB youth (D’Augelli 2005), and future research should better explore how parental knowledge operates in these instances. In addition, many of the articles did not control for other potentially important explanatory factors and may be affected by left out variable error (Mauro 1990).

Not surprisingly, parent–child relationships characterized by closeness and support again emerged as having a protective association with suicide among LGB youth (Friedman et al. 2006; Needham and Austin 2010; Proctor and Groze 1994; Resnick et al. 1997; Teasdale and Bradley-Engen 2010). For example, in a statewide school-based sample of adolescents, Eisenberg and Resnick (2006) found that family connectedness was negatively associated with suicide and accounted for a greater amount of variance in suicidal behavior than sexual orientation or any other protective factor. The strongest support was offered by the single prospective study, which found that perceived parental caring was negatively associated with suicidal tendencies for LGB youth and partially mediated the association between sexual orientation and suicidal tendencies in a nationally representative sample of LGB youth (Teasdale and Bradley-Engen 2010).

Taken together, there was modest support that parental knowledge and negative parental responses were important correlates of suicidal thoughts and attempts among LGB youth and adolescents. However, prospective research is needed to clarify the precise relationship between parental knowledge and suicidality among LGB youth. Stronger support was offered for the benefits of having a supportive, connected, and caring parent–child relationship. As with other health domains, the findings are limited by the use of convenience sampling strategies (n = 13), limited prospective research (n = 1), and limited attention to ethnic minority and rural youth.

Discussion

Our purpose was to assess the current state of knowledge on parental influences on LGB youth health. In total, we identified 31 articles that met our inclusion criteria. Overall, we find support that parents can be an important influence on LGB youth’s health and well-being. However, there were notable limitations. Below, we summarize the key findings and limitations and provide recommendations on how to improve the research base.

A key limitation in the extant literature is the reliance on convenience samples of LGB youth recruited via LGB-serving organizations or social venues or via snowball sampling techniques. In addition, few studies reported data on participation rates and refusal bias. As such, it is difficult to generalize many of the findings to the broader population of LGB young people. A number of researchers have suggested that maladaptive health behaviors and outcomes may be overestimated by the use of convenience samples (Binson et al. 2007; Savin-Williams 1998). In addition, it has been suggested that youth recruited from LGB-serving organizations are more open about their sexual orientation (Hershberger and D’Augelli 1995; Hershberger et al. 1997) and experience higher rates of victimization (Elze 2003). LGB populations have long been considered difficult to reach (Binson et al. 2007), and LGB youth may be especially difficult as many are still developing their sexual orientation (Savin-Williams 1998). Thus, though potentially difficult, there are probability sampling methods that can be utilized to improve the methodological rigor of scholarship in this domain (for a review of available methods, see Binson et al. 2007). Future research should employ these whenever possible as this will considerably strengthen the research base.

A second limitation addresses the dearth of prospective research with LGB youth. Of the 31 articles, only three presented longitudinal findings, and all of these were based on Add Health data. Future research should continue to maximize Add Health. As a nationally representative dataset, it is one of the few probability samples of LGB young people that enables researchers to explore parental influences in both adolescence and adulthood. At the same time, Add Health is unable to answer key questions, such as how responses to a child’s sexual orientation are related to LGB youths’ health (such questions were not included in the survey). Thus, additional prospective studies that can clarify the temporal relationships observed in cross-sectional research as well as the precise pathways of influence through which parents shape LGB youth health would be a welcome addition to the literature.

Our results suggest that research needs to better attend to the diversity among LGB youth. Although many articles included ethnic minority youth, the majority of sampled youth were White, with Native American, Asian American/Pacific Islander, and bi and multiracial youth being particularly understudied. This area of research is particularly important as studies suggest that race/ethnicity and cultural and familial values can influence the extent to which young people come out to their parents (Grov et al. 2006) and important others (Rosario et al. 2004), parents’ responses to their child’s sexual orientation (Garofalo et al. 2008), and youths’ responses to dynamics within the parent–child relationship (Parke et al. 2004). For example, families who embody more traditional values, such as speaking a language other than English in the home, valuing religion, and valuing the importance of marriage and having children, may be less accepting of their child’s sexual orientation (Newman and Muzzonigro 1993). Some research also has found that White youth are more likely to disclose their sexual orientation to their parents than are Latino and African American youth (Garofalo et al. 2008; Grov et al. 2006). At the same time, there were no differences in parental/caregiver rejection between Whites and Latinos observed by Ryan et al. (2009). Clearly, this is an area that deserves more focused attention and is likely to be particularly important in the context of developing tailored interventions for diverse groups of LGB youth.

Two other underexplored areas were the extent to which parental influences differed as a function of youth sexual orientation and area of geographical residence. In the present review, rural youth were underrepresented in most health areas. In one of the only studies exploring this potential difference, Savin-Williams (1989a) found that maternal knowledge and contact had different associations for youth living in rural versus urban areas. This is an important area of research to address as access to supportive services and resources for LGB youth and their families are likely to differ for youth residing in rural, suburban, and urban areas, with urban areas likely having a higher concentration of LGB-focused social venues, neighborhoods, and community-based organizations. In addition, empirical support for attending to the importance of sexual orientation was offered by Needham and Austin (2010), who observed significant mean differences in substance use and suicidal thoughts between gays, lesbians, and bisexuals, and that parental support operated as a mediator for some groups of LGB youth but not others. Similar findings have been observed in previous research (for a review, see Elze 2005; Volpp 2010). Currently, the reasons underlying these differences are not well understood, making this a particularly fruitful area for additional research.

Across the five health areas, the majority focused on mental health and suicide, with limited attention to parental influences on sexual behavior, substance use, and experiences with violence and victimization among LGB youth. The focus on these two health outcomes may reflect concern over existing epidemiological data (Hershberger and D’Augelli 1995; Silenzio et al. 2007). However, the lack of research on substance abuse and sexual behavior is particularly notable because LGB youth experience health disparities in each area (Garofalo et al. 1998; Hall et al. 2008; Marshall et al. 2008) and because a large and strong body of evidence suggests that parents are important in both of these domains (Hill et al. 2005; Miller et al. 2001). The lack of parent-based research and interventions in the domains of sexual behavior and substance use is particularly troubling given that young MSM of color are more likely to be infected with HIV during adolescence and to have higher rates of undiagnosed HIV infection than their White MSM peers (Campsmith et al. 2010; Hall et al. 2008).

Our review also indicates that we know relatively little about the perspectives of parents of LGB youth. In order to protect LGB youth from potential harm, the majority of articles in the review sought waivers of parental consent. However, there is evidence to suggest that many LGB youth disclose their sexual orientation to at least one parent and that a cohort effect is occurring, such that more contemporary groups of LGB youth are coming out to their parents at younger ages (Savin-Williams 1998; Grov et al. 2006). Although youth reports are more consistent correlates of health behaviors, parent perspectives remain important. Indeed, effective intervention programs cannot be designed without a better understanding of parents’ needs. Because protecting youths’ safety is of utmost concern, researchers can adopt methods that seek to involve parents and protect youth. These include asking youth to nominate parents to whom they have disclosed their sexual orientation and documenting the characteristics of youth who agree to nominate and those who refuse. In a recent study, 48% of LGB youth under age 18 agreed to have a parent contacted to provide parental consent (Elze 2003). Notably, no parent/legal guardian refused consent (Elze 2003).

In addition, future research should explore how specific characteristics of parents may be related to youths’ health. For example, Savin-Williams (1989a) observed that both maternal knowledge and self-esteem were higher among LGB youth with younger mothers, and previous studies have suggested that younger parents are more accepting than are older parents. While we focused our review on parenting style and goal-directed parenting practices, there are other parenting characteristics and behaviors that deserve additional attention. These include factors such as parental mental health and substance use, which have implications for parenting, the parent–child relationship, and LGB young people’s health and well-being (Corliss et al. 2010; Schneider et al. 1989).

Despite the existing limitations in the research, there were some key trends with respect to the potential influence of parents on the health and well-being of LGB youth. Although the cross-sectional study designs and dearth of research in certain health areas make it difficult to draw firm conclusions, the overall pattern of results suggested that two dimensions of parenting are important: (a) parents’ knowledge of and responses to their child’s sexual orientation and (b) emotional qualities of the parent–child relationship, such as support, caring, and parent–child connectedness. Across separate articles with diverse samples of youth, parental rejection was found to have a negative association with all five targeted health areas. In contrast, numerous studies indicated that a parent–child relationship characterized by support, acceptance, and connectedness was generally associated with less risky behavior and improved health outcomes. Although most of the research examined direct relationships, a smaller number investigated parental support as a potential moderator or mediator. Overall, the findings lend support to the idea that parents can serve as a source of stress and a source of support, and future research is needed to better understand how these two dimensions of parenting operate with diverse groups of LGB youth.

At the same time, it is premature to conclude that these two dimensions of parenting are what matter most. Currently, they are important because they are the two dimensions of parenting that have been most studied. A notable gap in the extant literature is the lack of research on how parental influences identified in the broader empirical literature, such as parental monitoring or parent–child communication, shape LGB youth’s health. None of the included articles examined these aspects of parenting, which is a stark contrast to the larger body of research. In addition, no studies conducted a simultaneous analysis of how these two dimensions of parenting work together to influence youth health or if these dimensions of parenting remain significant when controlling for other parenting practices and behaviors. As a result, a number of important questions remain unanswered. For example, two of the most commonly studied and robust parental influences are parental monitoring and parent–child communication. Research suggests that both practices are most effective when there is an open and mutual exchange of information between parents and youth (Kerr et al. 1999; Stattin and Kerr 2000) and when young people perceive that their parents are trustworthy and looking out for their best interests (Guilamo-Ramos et al. 2006). Given these factors, how then does parental monitoring operate in families where parents express ambivalence or rejection of their child’s sexual orientation? Can parents be effective monitors and communicators in these situations? These questions are not trivial and have important theoretical and applied implications. Future research should explore these areas, which will help to bring LGB research in line with the rigor, breadth, and complexity of scholarship on parental influences among the general youth population.

Across the articles, there was a trend to focus on negative parental influences as opposed to the mechanisms through which parents might be able to positively influence youths’ health. In some ways, this trend is consistent with the broader research on LGB youth, which has tended to focus on risk and pathology (Savin-Williams 2001). However, there is a clear need to identify protective mechanisms in the family that can buffer LGB youth from negative responses to their sexual orientation. Although strained parent–child relationships exist for many LGB youth, there is a strong public health imperative to identify mechanisms that can enable parents to support their child’s health, even in families where parents struggle to accept their child’s sexual orientation.

As with any study, the findings must be interpreted within the context of existing limitations. Because we focused specifically on parental influences, we excluded studies that examined the broader family system. Parents are one part of families and future research should examine how the influence of siblings, grandparents, or extended family members can be utilized to support LGB youth. In addition, we were not able to examine if parental influences differed between different types of parents, e.g., biological, stepparents, or legal guardians. Future research should examine the extent to which different types of family configurations are important for understanding parental influences on LGB youth. Our focus on parents also meant that we excluded other important ecological contexts, such as schools. Recent studies indicate that schools with LGB support groups, supportive and caring adults, and other supportive services are associated with lower rates of victimization and suicidality among LGB youth (Goodenow et al. 2006). Given that schools and families are two of the most important developmental contexts for young people, future research should explore how schools can support parents in their own efforts to keep their child safe from harm or potentially buffer youth from negative parental influences. Finally, although we excluded transgender youth from the review, we believe that parents are important for transgender youth and encourage researchers to better explore parental influences for this group of young people.

Although we did not identify any existing interventions, the overall pattern of results suggests that parent-based interventions are not necessarily contraindicated by the extant data. That is, among LGB youth who have disclosed their sexual orientation to their parents, there are important research and practice opportunities. Here, we join a growing chorus of prevention scientists who are calling for greater attention to the protective role of parents in the lives of LGB young people (D’Augelli 2005; Garofalo et al. 2008; Horn et al. 2009; Ryan et al. 2009; Savin-Williams 1998). Given that the majority of parents want their children to develop into healthy and productive adults, a careful consideration of the protective role of parents among LGB youth is warranted and represents a significant conceptual shift in the current public health literature.

Acknowledgments

This review was supported through funds from the Centers for Disease Control and Prevention, Division of Adolescent and School Health. The support was made possible through the Parenting Synthesis Project at CDC, DASH. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC.

Copyright information

© Springer Science+Business Media, LLC 2010

Authors and Affiliations

  • Alida Bouris
    • 1
  • Vincent Guilamo-Ramos
    • 2
  • Angela Pickard
    • 3
  • Chengshi Shiu
    • 1
  • Penny S. Loosier
    • 4
  • Patricia Dittus
    • 4
  • Kari Gloppen
    • 4
  • J. Michael Waldmiller
    • 4
  1. 1.University of Chicago School of Social Service AdministrationChicagoUSA
  2. 2.New York University Silver School of Social WorkNew YorkUSA
  3. 3.University of Toronto Dalla Lana School of Public HealthTorontoCanada
  4. 4.Centers for Disease Control and PreventionAtlantaUSA

Personalised recommendations