Parent-Initiated Sexual Orientation Change Efforts With LGBT Adolescents: Implications for Young Adult Mental Health and Adjustment

Abstract

Studies of adults who experienced sexual orientation change efforts (SOCE) have documented a range of health risks. To date, there is little research on SOCE among adolescents and no known studies of parents’ role related to SOCE with adolescents. In a cross-sectional study of 245 LGBT White and Latino young adults (ages 21–25), we measured parent-initiated SOCE during adolescence and its relationship to mental health and adjustment in young adulthood. Measures include being sent to therapists and religious leaders for conversion interventions as well as parental/caregiver efforts to change their child’s sexual orientation during adolescence. Attempts by parents/caregivers and being sent to therapists and religious leaders for conversion interventions were associated with depression, suicidal thoughts, suicidal attempts, less educational attainment, and less weekly income. Associations between SOCE, health, and adjustment were much stronger and more frequent for those reporting both attempts by parents and being sent to therapists and religious leaders, underscoring the need for parental education and guidance.

The American Psychiatric Association removed homosexuality from its diagnostic manual as a mental disorder more than four decades ago, yet efforts to change sexual orientation, often referred to as “conversion” or “reparative” therapy, continue to be practiced by some mental health providers, clergy, and religious leaders (APA Task Force on Appropriate Therapeutic Reponses to Sexual Orientation, 2009; Substance Abuse, Mental Health Services Administration, 2015). Although research on adult populations has documented harmful effects of sexual orientation change efforts (SOCE), no studies have examined SOCE among adolescents (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009). Yet because some people believe that homosexuality can be changed or “cured,” some parents engage in efforts to change their child’s sexual orientation, and some may seek professional therapies for a child’s same-sex sexual orientation. In this study we consider the health and adjustment of a sample of lesbian, gay, bisexual, and transgender (LGBT)¹ young adults in association with retrospective reports of efforts by their parents to change their sexual orientation during adolescence.

Existing research and field consensus

SOCE continues to be practiced despite a lack of credible evidence of effectiveness, reported harm from a range of studies on SOCE with adults (see APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009; SAMHSA, 2015), and increased adoption of practice guidance from major professional associations that caution against SOCE.² In one controversial study, 200 individuals who reported some change from homosexual to heterosexual following therapy were examined (Spitzer, 2003). The majority reported some minimal change from a homosexual to a heterosexual orientation; complete sexual orientation change was rare. The study received a great deal of attention and criticism for methodological limitations that included sample recruitment bias and problems in measurement and statistical reporting (see Drescher & Zucker, 2006 for a comprehensive review of the critiques of this study; the author later retracted the study). A review of 28 empirically based studies that have examined the use of these therapies strongly criticized the body of literature for multiple significant methodological flaws (see Serovich et al., 2008).

By the 1990s a wide range of major professional associations in the United States adopted position statements that supported affirmative care for lesbian, gay, and bisexual (LGB) clients and patients, and in the same time period several of them published statements that opposed efforts to change an individual’s sexual orientation (e.g., American Academy of Pediatrics, 1993; American Psychiatric Association, 1994; American Psychological Association, 1998; National Association of Social Workers, 1992). Despite these professional statements, some providers have continued to engage in SOCE with adults and adolescents, and the American Psychological Association (APA) convened a task force in 2007 to conduct a systematic review of peer-reviewed studies related to SOCE. The task force report concluded that published studies making claims that sexual orientation had been changed were methodologically unsound (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009). Moreover, the report noted that SOCE were unlikely to be successful and involved risk of harm. Specifically, studies of SOCE with adults (e.g., Shidlo & Schroeder, 2002) have reported a range of negative outcomes, including depression, anxiety, self-hatred, low self-esteem, isolation, and suicidality (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009).

Adolescents, parents, and SOCE

At the time of the APA report, no studies were identified that focused on sexual orientation change efforts among adolescents (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009); nevertheless, several organizations continued to market the effectiveness of sexual orientation change efforts for youth (see Ryan & Rivers, 2003). As the Group for the Advancement of Psychiatry—a policy organization that provides guidance for the psychiatric profession—has noted, “Despite … changes in scientific thinking in the last two decades, social and religious conservatives have advanced their own illness/behavior model of homosexuality [which] maintains that homosexuality is not inborn and that variations of long disproven theories of homosexuality’s etiology can serve as a basis for offering conversion therapies” (Drescher et al., 2016, p. 8).

Understanding adolescent experiences is especially important, particularly since SOCE with minors raises distinct ethical concerns (Hicks, 1999; Substance Abuse and Mental Health Services Administration, 2015). These include determining what constitutes appropriate consent, the potential for pressure from parents and other authority figures, the minor’s dependence on adults for emotional and financial support, and the lack of information regarding the impact of SOCE on their future health and wellbeing.

Concerned parents who believe that being lesbian, gay, or bisexual (LGB) is wrong and that an individual’s sexual orientation can be changed may engage in rejecting behaviors, such as trying to change their child’s sexual orientation; excluding them from family events and activities to discourage, deny, or minimize their identity; or using religion to prevent or change their sexual orientation (e.g., Ryan, Huebner, Diaz, & Sanchez, 2009). These parental behaviors are typically motivated by concern and represent efforts to try to help their child “fit in,” to be accepted by others, to conform with religious values and beliefs, and to meet parental expectations (Morrow & Beckstead, 2004; Ryan et al., 2009; Ryan & Rees, 2012; SAMHSA, 2014). Moreover, such efforts are based on a belief that homosexuality is a mental illness or developmental disorder that needs to be corrected or cured. Yet SOCE are at odds with mainstream understandings of human development and professional standards of care, which hold that LGB identities are normative and that social stigma and minority stress contribute to negative health outcomes and self-hate (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009; Substance Abuse and Mental Health Services Administration, 2015).

There is growing concern that SOCE has continued to be practiced despite serious ethical conflicts and potentially harmful effects (APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation, 2009). An analysis by the Williams Institute estimated that nearly 700,000 U.S. LGBT adults have received SOCE conversion therapy interventions, including 350,000 LGBT adults who received SOCE interventions as adolescents (Mallory, Brown & Conron, 2018). This concern led legal advocates in the United States to introduce legislation to prevent SOCE among licensed practitioners, an approach that has been adopted in 10 U.S. states and a growing number of jurisdictions and that has sought to inform families, the public, practitioners and religious leaders of the impact of such practices (Drescher, 2013; Movement Advancement Project, 2018). Although these laws appear to have raised awareness and informed public perceptions and responses (Ames, 2015), they do not prevent SOCE in families or by unlicensed practitioners, clergy, and others.

The U.S. Substance Abuse and Mental Health Services Administration asked the APA to convene a scientific advisory panel of researchers and practitioners who were experts in the field to review existing research, professional policies, and clinical guidelines to develop consensus recommendations related to the ethical and scientific foundations of conversion therapy with minors (Substance Abuse and Mental Health Services Administration, 2015). Concurrently, the Obama administration called for an end to conversion therapy of minors, citing, in particular, the importance of family support for LGBT young people (Jarrett, 2015). Most recently, in March 2018 the European parliament passed a resolution condemning the practice and urging member nations to ban SOCE.

The current study

Historically, SOCE research has focused on adults. Decades ago, Gonsiorek theorized that the experience of SOCE during adolescence can “contribute to negative self-esteem and mental health problems” (Gonsiorek, 1988, p. 116), yet there are no known studies of the link between such interventions and the health and wellbeing of lesbian, gay, bisexual, and transgender (LGBT) young people, particularly SOCE efforts carried out both by parents and caregivers, as well as by practitioners and religious leaders.

To our knowledge, we present the first study to examine young adults’ retrospective reports of parent-initiated efforts to change their sexual orientation during adolescence, and the associations between these experiences and young adult mental health and adjustment. The two goals of this study include: (1) to identify demographic and family characteristics that are associated with parent-initiated attempts to change a child’s sexual orientation during adolescence, and (2) to examine associations among these parent-initiated attempts in adolescence with a range of indicators of young adult health and adjustment.

Method

The sample included 245 participants who self-identified as LGBT. Participants were recruited from local bars, clubs, and community agencies that serve this population in a 100-mile radius of the research center. Screening procedures were used to select participants into the study based on the following criteria: age (21–25); ethnicity (White, Latino, or Latino mixed); self-identification as LGBT during adolescence; being open about LGBT status to at least one parent or guardian during adolescence; and having lived with at least one parent or guardian during adolescence at least part-time. The survey was administered in both English and Spanish, and it was available in either computer-assisted or paper-and-pencil format. The study protocol was approved by the university’s institutional review board.

Sample

Of the 245 participants, 46.5% were male, 44.9% were female, and 8.6% were transgender. The majority of participants identified as gay (42.5%), 27.8% as lesbian, 13.1% as bisexual, and 16.7% as other (e.g., queer, dyke, homosexual). Approximately one half of the sample identified as Latino (51.4%), and the other 48.6% identified as White, non-Latino. In addition, 18.78% of the respondents were immigrants to the United States. The age of the participants ranged from 21 to 25 years (M = 22.8, SD = 1.4). Family of origin socioeconomic status was assessed retrospectively (1 = both parents in unskilled positions or unemployed to 16 = both parents in professional positions; M = 6.75, SD = 4.77).

Measures

Parent-initiated efforts to change youths’ sexual orientation

Participants responded to two items that assessed past parental and caregiver-initiated efforts to change the youths’ sexual orientation. The first item asked: “Between ages 13 and 19, how often did any of your parents/caregivers try to change your sexual orientation (i.e., to make you straight)?” (0 = never [49.64%]; 1 = ever [53.06%]). A second item asked: “Between ages 13 and 19, how often did any of your parents/caregivers take you to a therapist or religious leader to cure, treat, or change your sexual orientation?” (0 = never [65.71%]; 1 = ever [34.29%]). We created a single measure with three categories that identifies the severity of parent-initiated attempts to change youths’ sexual orientation. The three categories include: (1) no attempt to change sexual orientation (n = 109; 41.63%), (2) parent and caregiver-initiated attempt to change sexual orientation without external conversion efforts (n = 52; 21.22%), and (3) parent and caregiver-initiated attempt to change sexual orientation with external conversion efforts (n = 78; 31.84%). Six participants who reported conversion efforts but not parental attempts to change sexual orientation were dropped from the current study, for a total analytic sample of 239 participants.

Young adult health and adjustment

Indicators of mental health and adjustment assessed included suicidal ideation, lifetime suicidal attempts, depression, self-esteem, and life satisfaction. Suicidal ideation was assessed by one item: “During the past six months, did you have any thoughts of ending your life?” (0 = no, 3 = many times). Lifetime suicidal attempts were assessed by one item: “Have you ever, at any point in your life, attempted taking your own life?” (0 = no, 1 = yes). Depression was measured by the 20-item CES-D (Radloff, 1977, 1991). Two dichotomized cut-off scores were also used: a clinical cut-off score (≥16) and a prescription intervention cut-off score (≥22). Self-esteem was measured by Rosenberg’s 6-item self-esteem scale (Rosenberg, 1979). Life satisfaction was measured by an 8-item scale (e.g., “At the present time, how satisfied are you with your living situation?”). Social support was measured by the 12-item Multidimensional Scale of Perceived Social Support (Zimet, Dahlem, Zimet, & Farley, 1988).

Behavioral health risk indicators included substance use and abuse and engagement in risky sexual activities. Binge drinking (or heavy alcohol use) was assessed by two items that measured the frequency of drinking and the number of drinks per occasion (0 = less than 1–2 times per week and less than 3 drinks per occasion; 1 = 1–2 times per week or more and more than 3 drinks per occasion). Substance abuse problems were assessed by four items (e.g., “In the past five years, have you had problems with the law because of your alcohol or drug use?”) and were dichotomized to represent ever having problems versus never having problems. Risky sexual behavior was assessed in six ways: unprotected sex during the last 6 months (0 = no, 1 = yes), unprotected sex with a casual or HIV positive partner during the last 6 months (0 = no, 1 = yes), unprotected sex during last sexual encounter (0 = no, 1 = yes), unprotected casual sex during last sexual encounter (0 = no, 1 = yes), ever been diagnosed with a sexually transmitted disease (0 = never, 1 = ever), and one item that assessed HIV risk (“In the last six months, were you ever at risk for being infected with or transmitting HIV?”; 0 = no, 1 = yes).

Finally, two indicators of young adult socioeconomic status were assessed: current monthly income and educational attainment. Current weekly income as assessed by one item: “What is your personal weekly income (after taxes, unemployment, social security, etc.)?” (1 = less than $100, 7 = more than $2000). Educational attainment was assessed by one item: “What is the highest level of education you have completed?” (1 = less than elementary school, 7= postgraduate).

Demographic and family characteristics

Adolescent gender nonconformity and family religiosity were included as possible characteristics that may predict whether or not parents/caregivers attempted to change the participant’s sexual orientation during adolescence. Adolescent gender nonconformity was measured by one item: “On a scale from 1–9, where 1 is extremely feminine and 9 is extremely masculine, how would you describe yourself when you were a teenager (age 13–19)?” This item was reverse coded for males such that a higher score is representative of more nonconformity to gender norms (M = 4.40, SD = 1.87). Family religiosity was measured by one item: “How religious or spiritual was your family while you were growing up?” (0 = not at all, 3 = extremely; M = 1.35, SD = 0.91).

Plan of analysis

First, demographic and family characteristics were included in a multinomial logistic regression to predict the likelihood of a participant experiencing parent-initiated attempts to change their sexual orientation during adolescence without external conversion intervention efforts (= 1) and parental attempts to change sexual orientation with external conversion efforts (= 2) compared to no attempts (= 0). Second, to understand the associations among parent-initiated attempts to change the participant’s sexual orientation during adolescence with young adult health and wellbeing, we used logistic regressions for dichotomous outcomes and multiple linear regression for continuous outcomes, including known covariates for the outcomes of interest (Ryan et al., 2009). To minimize exclusion of participants due to missing data and to maximize statistical power, we used PRELIS, a component of LISREL, to impute missing data (total <5%; Graham, Cumsille, & Elek-Fisk, 2003) using all numeric variables in an expectation maximization algorithm for imputation. All continuous variables were checked for assumptions of normality; the depression measure was significantly skewed, but after a square-root transformation the items met assumptions of normality. Finally, we conducted linear trend analyses for study outcomes across the three groups of participants based on no attempts, parent-initiated attempts, and parent-initiated attempts with external conversion efforts.

Results

Similar background characteristics predicted both types of parent-initiated SOCE (see Table 1). Notably, there were no differences in reports of SOCE based on gender, sexual identity (bisexual or queer), or ethnicity. However, adolescents who grew up in religious families were more likely to experience SOCE (with and without external conversion efforts). Higher family of origin socioeconomic status was also associated with fewer parent-initiated SOCE (with and without conversion efforts). Additionally, participants who were not born in the United States and who reported more gender nonconformity during adolescence were more likely to experience parent-initiated attempts to change with external conversion efforts.

Table 1.

Demographic and family characteristics predicting parent/caregiver-initiated sexual orientation change efforts.

	Parent-initiated SOCE	Parent-initiated SOCE with external conversion efforts
Female (Ref = male)	1.62 (0.76–3.46)	0.94 (0.46–1.92)
Transgender (Ref = male)	2.30 (0.40–13.14)	1.93 (0.44–8.47)
Bisexual (Ref = gay/lesbian)	0.80 (0.30–2.17)	0.40 (0.13–1.23)
Queer (Ref = gay/lesbian)	0.49 (0.14–1.74)	1.24 (0.46–3.34)
White, non-Latino (Ref = Latino)	0.86 (0.39–1.90)	1.51 (0.70–3.23)
Immigrant (Ref = U.S. native)	1.98 (0.67–5.90)	6.47 (2.43–17.23)***
Family of origin SES	0.85 (0.78–0.93)***	0.88 (0.81–0.95)***
Adolescent gender nonconformity	1.18 (0.96–1.45)	1.27 (1.05–1.54)*
Family religiosity	1.72 (1.13–2.61)*	1.88 (1.28–2.76)**

N = 239. Ref = reference group. Adjusted odds ratios and 95% confidence intervals from a multinomial logistic regression are shown. The reference category for the model was “neither change efforts nor conversion efforts.”

^***p < .001.

^**p < .01.

^*p < .05.

Table 2 displays the results of logistic and linear regressions predicting young adult health and adjustment based on reports of parent-initiated SOCE during adolescence (both with and without external conversion efforts). Both levels of parent-initiated attempts to change participant’s sexual orientation during adolescence were associated with more negative mental health problems for young adults. Specifically, those who experienced SOCE were more likely to have suicidal thoughts (although only for those who reported SOCE with external conversion efforts) and to report suicidal attempts and higher levels of depression. Participants who experienced SOCE had lower life satisfaction and less social support in young adulthood. Parental-initiated SOCE in adolescence were not associated with self-esteem, substance use or abuse, or risky sexual behavior. Finally, parent-initiated SOCE during adolescence were associated with lower young adult socioeconomic status: less educational attainment and less weekly income (although only for those who experienced attempts to change with external conversion efforts).

Table 2.

Parent/caregiver-initiated sexual orientation change efforts predicting young adult outcomes.

	Parent-initiated SOCE	Parent-initiated SOCE with external conversion efforts
Mental Health
Suicidal ideation (continuous)	0.13	0.27***
Suicidal attempt (1 = ever)	3.08 (1.39–6.83)**	5.07 (2.38–10.79)***
Depression – Clinical cut-off score (≥16)	2.20 (1.02–4.73)*	3.92 (1.92–8.00)***
Depression – Prescription intervention cut-off score (≥22)	1.94 (0.82–4.57)	3.63 (1.67–7.87)**
Depression (continuous)	0.15*	0.30***
Self-esteem (continuous)	−0.13	−0.13
Life satisfaction (continuous)	−0.19**	−0.34***
Social support (continuous)	−0.26***	−0.45***
Substance Use/Abuse
Binge drinking (1 = yes)	0.90 (0.42–1.93)	1.01 (0.50–2.03)
Substance abuse problems (1 = yes)	0.87 (0.42–1.82)	1.70 (0.84–3.44)
Sexual Risk Behavior
Unprotected sex during last 6 months (1 = yes)	1.61 (0.70–3.72)	2.05 (0.91–4.59)
Unprotected sex with casual or HIV + partner last 6 months (1 = yes)	0.91 (0.36–2.30)	2.09 (0.91–4.78)
Unprotected sex at last intercourse (1 = yes)	0.90 (0.43–1.87)	1.23 (0.62–2.45)
Unprotected casual sex at last intercourse (1 = yes)	1.01 (0.41–2.49)	1.11 (0.48–2.58)
STD diagnosis (1 = ever)	0.79 (0.33–1.91)	1.36 (0.62–2.99)
HIV risk in last 6 months (1 = yes)	0.74 (0.31–1.74)	1.06 (0.50–2.26)
Current Socioeconomic Status
Educational attainment (continuous)	−0.15*	−0.32***
Current weekly income (continuous)	−0.12	−0.27***

N = 239. Adjusted odds ratios and 95% confidence intervals are shown for dichotomous outcomes and standardized beta coefficients are shown for continuous outcomes. All analyses controlled for gender, sexual orientation, ethnicity, immigrant status, family of origin socioeconomic status, adolescent gender nonconformity, and family of origin religiosity.

^***p < .001.

^**p < .01.

^*p < .05.

Differences across the three groups defined by parent-initiated SOCE are presented in Table 3. Trend analyses confirmed that parental attempts to change adolescents’ sexual orientation are significantly associated with negative health outcomes in young adulthood, and that those problems are worse for young adults who experienced SOCE that included external conversion efforts during adolescence. This pattern of results emerged as statistically significant for 12 of the 18 outcomes tested, including significant findings for all outcomes related to mental health and socioeconomic status.

Table 3.

Trend effects related to parent/caregiver-initiated sexual orientation change efforts predicting young adult health outcomes.

	No SOCE (n = 109)	Parent-Initiated SOCE (n = 52)	Parent-Initiated SOCE with External Conversion Efforts (n = 78)	Group difference (X2; F)
Mental Health
Suicidal ideation (continuous)	.17	.38	.57	***
Suicidal attempt (1 = ever)	22.0 %	48.1 %	62.8 %	***
Depression – Clinical cut-off score (≥16)	26.6 %	46.2 %	65.4 %	***
Depression – Prescription intervention cut-off score (≥22)	15.6 %	32.7 %	52.3 %	***
Depression (continuous)	9.21	12.99	16.10	***
Self-esteem (continuous)	2.88	2.74	2.72	**
Life satisfaction (continuous)	3.05	2.78	2.61	***
Social support (continuous)	4.18	3.66	3.31	***
Substance Use/Abuse
Binge drinking (1 = yes)	42.2 %	36.5 %	41.3 %	NS
Substance abuse problems (1 = yes)	49.5 %	50.0 %	66.7 %	*
Sexual Risk Behavior
Unprotected sex during last 6 months (1 = yes)	28.4 %	36.5 %	42.3 %	*
Unprotected sex with casual or HIV + partner last 6 months (1 = yes)	22.0 %	21.2 %	38.5 %	*
Unprotected sex at last intercourse (1 = yes)	49.5 %	53.9 %	59.0 %	NS
Unprotected casual sex at last intercourse (1 = yes)	15.6 %	23.1 %	25.6 %	NS
STD diagnosis (1 = ever)	24.8 %	21.2 %	30.8 %	NS
HIV risk in last 6 months (1 = yes)	28.4 %	25.0 %	37.2 %	NS
Current Socioeconomic Status
Educational attainment (continuous)	5.19	4.65	4.26	***
Current weekly income (continuous)	2.73	2.31	2.03	***

Six participants who reported conversion efforts but not parent attempts are excluded. Percentages are shown for dichotomous outcomes with chi-square significance levels, and average scores are shown for continues outcomes with ANOVA F significance levels.

^***p < .001.

^**p < .01.

^*p < .05.

Discussion

Results from this study clearly document that parent/caregiver efforts to change an adolescent’s sexual orientation are associated with multiple indicators of poor health and adjustment in young adulthood. The negative associations were markedly stronger for participants who experienced both parental attempts to change their sexual orientation, coupled with efforts to send the adolescent to a therapist or religious leader to change their sexual orientation (strategies often called “conversion” or “reparative” therapy). In this sample of LGBT young adults, more than half reported some form of attempt by their parents and caregivers to change their sexual orientation during adolescence. With the exception of high-risk sexual behavior and substance abuse, attempts to change sexual orientation during adolescence were associated with elevated young adult depressive symptoms and suicidal behavior, and with lower levels of young adult life satisfaction, social support, and socioeconomic status. Thus SOCE is associated with multiple domains of functioning that affect self-care, wellbeing, and adjustment.

The results of this study point to a number of factors that impact practice and provision of appropriate care. Family religiosity was strongly linked to parental attempts to change sexual orientation. In a related study, families that were highly religious were least likely to accept their LGBT children (Ryan, Russell, Huebner, Diaz, & Sanchez, 2010). Religiously conservative families often have misinformation about sexual orientation and gender identity and need accurate information to help support their LGBT children in the context of their values and beliefs (for guidance see Ryan & Rees, 2012; Substance Abuse Mental Health & Services Administration, 2014). Moreover, parents and caregivers often conflate sexual orientation with gender expression. Discomfort with gender nonconformity may be at the root of much of parents’ and caregivers’ motivations for SOCE: in the current study, gender nonconforming youth were more likely to experience attempts to change their sexual orientation through conversion therapy with therapists and religious leaders. Further, our results show that immigrant parents are more likely to try to change their children’s sexual orientation by sending them for clinical or religious intervention.

Related research has found that SOCE typically happens in the context of other family rejecting behaviors that contribute to health risks in young adulthood (Ryan et al., 2009). Parents, caregivers and others who provide support for LGBT children and adolescents need to understand that family rejection encompasses a wide range of behaviors, and education is critical for families, providers, and religious leaders on the relationship between family rejection and acceptance with health and wellbeing for LGBT young people (Ryan, 2009; Ryan & Chen-Hayes, 2013; Ryan et al., 2010; Substance Abuse and Mental Health Services Administration, 2015; Substance Abuse Mental Health & Services Administration, 2014).

Studies on responses of parents and caregivers with LGBT children indicate that parents’ reactions are motivated by a number of concerns, which include helping their child “fit in” to their family and cultural world, responding to religious and cultural values, keeping their families together, and trying to protect their LGBT child from harm (Maslowe & Yarhouse, 2015; Ryan, 2009; Substance Abuse Mental Health & Services Administration, 2014). In other words, parents are typically motivated by doing what they think is best for their child. Nonetheless, our study did not directly examine the motives of the parents of study participants. However, these findings reinforce the critical need for culturally appropriate family education and guidance on sexual orientation and gender identity and expression, the harmful effects of family rejecting behaviors, including SOCE, and the need for supporting their LGBT children, even in the context of parental and familial discomfort and religious conflict.

There are several limitations of this study. First, study inclusion criteria called for current identification as LGBT; it is likely that this inclusion criterion excludes persons who are dissatisfied with their LGBT identity, or persons who had identified as LGBT during adolescence but not at the time of the study. Thus we acknowledge that we did not include young people whose sexual orientation may be more fluid (e.g., sexual orientation in adolescence not consistent with sexual orientation in young adulthood). Second, although the study included a measure of family religiosity, there is no measure of specific religious affiliation, a factor that might be a further predictor of the role of parents in SOCE of their children. Third, the design is retrospective, and thus causal claims cannot be made. We cannot rule out the possibility that those who were most maladjusted as young adults retrospectively attribute parental behaviors during adolescence as attempts at changing their sexual orientation; we also cannot rule out the possibility that well-adjusted LGBT young adults may be less likely to recall experiences related SOCE. However, we note that the face validity of the specific measures is compelling: the alternatives are less plausible than the explanation that sexual orientation change attempts would likely undermine health and wellbeing.

Most attention to SOCE has focused on the ethics of professional practice and recent efforts to end such practice through legislation. This study highlights the crucial role parents play in SOCE—either directly themselves or through sending their children to therapists or religious leaders. Results point to the need for multicultural and faith-based family education resources and approaches to help parents and caregivers learn how to support their LGBT children in the context of their family, cultural, and religious values (see, for example, Kleiman & Ryan, 2013; Ryan, 2009; Ryan & Rees, 2012). In addition to supporting families and educating religious leaders and congregations, legislative and professional regulatory efforts to end SOCE therapies are important for raising awareness about and preventing a contraindicated practice that contributes to health risks, and for changing negative attitudes and bias regarding LGBT people.

Taken together, these findings provide a needed empirical framework for understanding the scope of SOCE in and outside of the home and the costs of sexual orientation change efforts directly from those individuals who are most affected—LGBT young people themselves. Historically, research and strategies to prevent SOCE have focused on mental health practitioners and much less on religious leaders, with limited awareness of the role of families in pressuring LGBT young people to change core identities. As indicated by this study, more attention is needed on family-based efforts to change a child’s sexual orientation and gender expression. Because LGBT youth cannot escape family rejecting behaviors (see, for example, Ryan, 2009; Ryan & Rees, 2012), approaches to prevent and ameliorate efforts to change a child’s sexual orientation and gender identity must include the broader social context that includes the home and social, cultural, and religious influences on families and caregivers to change or suppress a child’s sexual orientation and gender expression.