Psychotherapy for Unwanted Homosexual Attraction Among Youth

American College of Pediatricians – January 2016

ABSTRACT: Although there are no scientific studies which evaluate psychotherapy for unwanted homosexual attraction (UHA) among adolescents, there are four studies that examine “sexual orientation change efforts”(SOCE) among adults which have been referenced to support legislative efforts to ban minors from receiving psychotherapy for UHA. This review critically examines each of those four studies.  Pediatricians, mental health providers, educators, and policy makers need to know there is no evidence that psychotherapy for UHA is any more or less harmful than the use of psychotherapy to treat any other unwanted psychological or behavioral adaptation. Therefore, science does not support laws that prohibit minors with UHA from receiving psychotherapy in accordance with their personal goals and values.


In order to assess the claim that providing psychotherapy to minors with unwanted homosexual attraction (UHA) is substantially harmful, Dr. Christopher Rosik, past President of the Alliance for Therapeutic Choice, conducted a Medline and PsycARTICLES search of the medical and psychiatric literature.1 Medline and PsycARTICLES are the major medical and mental health databases utilized by the medical and psychological community. Both searches revealed that there is not a single study of youth who have received psychotherapy for UHA. Instead, all claims of harm to youth in the literature are based upon one of three categories of research: anecdotal accounts of harm experienced by adults who engaged in sexual orientation change efforts (SOCE), inferences from other research domains unrelated to psychotherapy for UHA (e.g., harms from family rejection of gay youth), and citations of the pronouncements on SOCE from professional mental health and medical associations. These various sources cite one another in an almost symbiotic manner that provides no objective information regarding youth who choose psychotherapy for UHA.

SOCE: an unscientific term

In 2012 the American Psychological Association published Guidelines for psychological practice with lesbian, gay, and bisexual clients.2 The third guideline (which is based upon a single adult study) states that sexual orientation change efforts have not been proven effective, and asserts that attempts to change sexual orientation “cause harm to many clients.”2 Accordingly, the guideline directs therapists to discourage patients with UHA from pursuing their goal of diminishing their homosexual attraction, and to offer those patients gay affirming therapy instead.

“Sexual Orientation Change Efforts” (SOCE) is a term coined by the American Psychological Association (APA) to replace terms such as reparative, ex-gay, change and conversion therapy. SOCE, however, is a dubious and problematic term. This is because “efforts” includes all forms of psychotherapy, 12 step programs, prayer meetings, unethical aversion therapies and everything in between. The following analogy illustrates why this is scientifically problematic. Imagine if physicians used the term “Alcoholic Change Efforts” (ACE) to describe all the possible ways alcoholics distressed by their unwanted alcohol attraction may pursue change. Some may enter psychodynamic therapy to learn and treat the underlying issues for which they self-medicate with alcohol; others may join Alcoholics Anonymous, some may join weekly prayer meetings for help, others may choose aversive pharmacologic therapy (e.g.: antabuse), and still others may engage in a combination of two or more of the above. All of these “efforts” attempt to work toward the goal of improved health and social function, but only one of these “efforts” is psychotherapy. Consequently, it would be inaccurate to analyze the individuals engaged in these various “efforts” in a single study as though they were receiving the same treatment and to then make claims regarding how effective or ineffective professional psychotherapy is for alcoholics. SOCE research, as will become evident, is rife with such ambiguity and is therefore fatally flawed.

An additional problem with the term SOCE is that the APA does not define what constitutes successful “change.” To continue with the previous analogy, some may argue that only alcoholics who attain a lifetime of abstinence are “changed.” Yet, some alcoholics will only be able to decrease their intake and dependence upon alcohol. Both groups of individuals will have experienced a healthful “change” though their endpoints differ. In other words, professional medical and psychiatric organizations should and often do recognize that successful change occurs on a continuum; with regard to psychotherapy for UHA, however, the APA fails to acknowledge  this reality.3

It is also crucial to understand that legislation put forth and passed by gay advocates solely bans minors with UHA from receiving professional psychotherapy. These laws leave minors with UHA no choice but to affirm what they perceive as a false sexual identity, or to pursue their desire for change under the direction of unlicensed individuals and/or religious ministries. Far from ensuring adolescent well-being, this legislation eliminates two of the greatest safeguards for patient health:(1) the right to informed consent and (2) the right to self-determination. Such hostility to patient rights is unprecedented within the mental health field and directly violates Principle E of the APA’s ethics code published in 2010. The code, entitled Respect for People’s Rights and Dignity, states “Psychologists respect the dignity and worth of all people, and the rights of individuals to privacy, confidentiality, and self-determination.”4

Since the 1970s, psychological and psychiatric professionals have increasingly deemed aversion therapies to be unethical and have therefore abandoned them without the need for judicial or legislative intervention.5 For example, the Alliance for Therapeutic Choice and Scientific Integrity published practice guidelines for therapists who assist clients with unwanted homosexual attractions.6 Aversion therapy would be excluded by Guideline 6, which states that, “Clinicians are encouraged to utilize accepted psychological approaches to psychotherapeutic interventions that minimize the risk of harm when applied to clients with unwanted same-sex attraction.” Aversion therapy violates the principle of this Guideline. Remarkably, gay activists in the state of Washington actually defeated bipartisan legislation which would have banned  aversion therapy for UHA because the legislation still allowed ethical (non-aversive) forms of psychotherapy for UHA. This opposition underscores all the more that it is the patient’s right to self-determination (specifically, his or her right to choose professional help with UHA) that is under attack, not an abusive therapeutic technique.

Prior to 2002, criticism of SOCE was based solely upon theoretical arguments and anecdotal reports.8 This changed with the publication of a survey by Drs. Ariel Shidlo and Michael Schroeder. The 2009 APA Task Force Report cited their work alone as definitive proof that “many are harmed by SOCE” and, based on it, concluded that all forms of SOCE should be discouraged.9  The reason anti-SOCE advocates have relied primarily upon this study is that, until recently, it was the only one to provide quantitative data. Three similar surveys have since been published. This statement reviews each of the four studies and closes with a summative discussion regarding psychotherapy for UHA.

Changing sexual orientation:  A consumer’s report by Drs. Shidlo and Schroeder (2002)10


Shidlo and Schroeder sought to provide empirical evidence on SOCE for individuals pursuing the goal of altering homosexual attractions. The researchers also investigated how individuals perceived their failure to change or their success in changing, and speculated how their survey results might impact ethical issues of SOCE. They studied a convenience sample of 202 individuals who reported pursuing a change in their sexual orientation. They found that 87% of participants described themselves as failing therapy and of experiencing some form of harm. Only 13% (26 participants) perceived themselves as having been successful. Twelve of the successful clients described themselves as “successful and struggling with behavior management techniques,” six identified as “successful and not struggling with behavior management techniques,” and eight described themselves as experiencing a “complete heterosexual shift.” Among the successful clients were also perceptions of benefits beyond a change in behavior. These included other psychological benefits such as a sense of hope, improved self-esteem, increased sense of belonging, improvement in social relationships with friends and family, and spiritual benefits. In contrast, among those who perceived the therapy to be harmful were reports of depression, suicidal ideation, decreased self-esteem, sexual dysfunction, and loss of social supports when entering and leaving the ex-gay community; some also perceived spiritual harm.


The authors acknowledge significant limitations in their study design from the outset. Specifically, they admit to potential researcher bias (they are both openly homosexual psychologists), recruitment bias (they specifically advertised for participants who had failed therapy), recall bias (most respondents received therapy years before the survey), and self-report bias as researchers lacked any objective validation of respondents’ claims and/or experience. The authors also did not differentiate results according to whether respondents had received “therapy” from trained mental health professionals, a religious ministry, another lay source, or multiple sources. Additionally, authors did not rule out pre-morbid psychological conditions including depression and suicidality. Therefore, it is possible that the reported episodes of depression, suicidality, and other distressing psychological symptoms pre-dated rather than resulted from SOCE.

To fully document the inherent sample bias, the authors included an appendix displaying the initial text used in participant recruitment, which was directed toward self-perceived treatment failures:

Have you gone through counseling or therapy where you were encouraged to become heterosexual or ex-gay? The National Lesbian and Gay Health Association wants to hear from you. The organization is conducting research for a project entitled “Homophobic Therapies: Documenting the Damage.” The NLGHA is conducting a survey of lesbians, gay men, and bisexuals who have been in counseling that tried to change their sexual orientation. They intend to use the results to inform the public about the often harmful effects of such therapies. Participation in the survey is confidential. Persons who are interested in responding can participate either through e-mail, by telephone, or in person. No record of your name, Internet address, or any other identifying information will be kept.10

After the initial 20 interviews, in which the authors received unexpected reports of positive outcomes, the recruitment verbiage was changed to be less biased against identifying positive outcomes. However, the authors continued to recruit subjects solely from pro-LGBQT (pro-gay-affirmative therapy) publications. Consequently, significant selection bias remained and was acknowledged by the authors themselves.


This study has anecdotal value only. The authors have documented that there are individuals who have negative experiences attempting to diminish UHA, and there are others who have positive experiences; nothing more. The authors themselves state forthrightly that the data they presented, “do not provide information on the incidence and prevalence of failure, success, harm, help, or ethical violations in conversion therapy” (p.250, emphasis in original).10  Given this admission within the body of the paper, it is dishonest for the APA and others to claim that this research proves unacceptable rates of failure or harmful outcomes for patients who pursue their informed choice to diminish UHA under the care of a licensed mental health provider.

Sexual reorientation therapy interventions by Drs. Flentje, Heck, & Cochran (2013)11


Flentje and colleagues set out to study “typical modalities and interventions” used to facilitate SOCE.  They surveyed 38 individuals who had gone through at least one “episode” of SOCE and later reclaimed a lesbian, gay or bisexual (LGB) identity.  According to the authors, the results revealed that frequently used interventions had a strong emphasis on religious practices, including negative messages about LGB individuals, and employed techniques that emphasized change over validation.  Some alleged unethical practices were also noted. Among the professional and policy recommendations the authors draw from their investigation is the endorsement of legal efforts to ban the option of psychotherapy from minors with UHA.


Sample bias

In addition to being an extremely small study with low statistical power (N=38), the sample composition was highly skewed toward males (n =31), Caucasians (n = 33), and those from a highly educated background (all but one subject having completed at least a 4-year college education). This calls into question the ability to generalize findings to individuals who are less educated, non-Whites, youth, and women.

Concerns for sample bias multiply when the authors detail the setting and type of counselor participants reported as providing their SOCE.  The majority of SOCE “episodes” (56.1%) were provided by religious or pastoral counselors.  Another 16.8% were administered by peer counselors.  Only 34.6% of SOCE “episodes” were actually provided by a licensed mental health professional.  The failure of this study to disentangle religious providers from licensed therapists is a serious limitation that makes it inappropriate to draw any definitive conclusions regarding professionally conducted SOCE.

Recruitment bias

Subjects were recruited through various “ex-ex-gay” listservs. “Ex-ex-gay” individuals are those who identified as “ex-gays” at some point during their SOCE and who at the time of the study once again identified as lesbian, gay, or bisexual (LGB). This is clearly a significant bias since persons who decide to reclaim an LGB identity following failed attempts to change their same-sex attractions and behaviors are not likely to look back on those attempts with particular favor. Moreover, participants rated themselves as being “exclusively homosexual” (n =22) or “predominately homosexual” (n = 16) both prior to engaging in SOCE and at the time of the study. This indicates the sample represented the most subjectively unalterable end of the same-sex attraction spectrum.

Recall bias and self-report bias

The authors acknowledge that participant reports were retrospective and that this may have impacted the accuracy of their accounts. It can be deduced from some of the statistics that some recollections are of SOCE that occurred at least 15 years prior to the survey. This study also suffers from self-report bias in that the authors had no way of objectively validating the participants’ claims.

Failure to account for potential pre-morbid psychopathology

Ten subjects reported having attempted suicide.  Of these, six subjects reported a suicide attempt prior to their SOCE, seven subjects reported 1 or 2 suicide attempts during SOCE, and only one participant indicated 2 suicide attempts following the conclusion of their SOCE. These findings suggest a significant portion of the sample was experiencing serious emotional distress prior to their SOCE, distress which cannot be definitively attributed to their SOCE experience in the absence of longitudinal data.

Failure to clearly identify type of provider involved in unethical modalities

The authors report that ethically questionable interventions occurred during 13 different courses of therapy reported by 10 different participants. They state that nine of these 13 episodes “…included a licensed or licensable professional as one of the providers of therapy” (emphasis added, p. 266).11 While the authors note in this section that the only instance of holding therapy was performed by an “ex-gay layperson” to whom the subject had been referred by his pastor, they do not specify who performed the aversive techniques in this section. The reader is left not knowing whether these were performed by licensed mental health professionals or someone else involved in the subjects’ care. The likelihood that these interventions were not provided by licensed mental health professionals but by laypeople is given credence by the authors’ statement in a previous section that no licensed therapist was described as utilizing aversion therapy.


The profound methodological flaws described above render the Flentje et al. nothing short of agenda driven research. No definitive claims about providing psychotherapy to adults or minors seeking to diminish homosexual attraction may be made based on this study.  In this regard it resembles the earlier research by Shidlo and Schroeder, whose methodological shortcomings it repeats, only this time accompanied by unjustified conclusions regarding harm, lack of benefit, and professional practice.

Sexual orientation change efforts among current or former LDS church members

by Drs. Dehlin, Galliher, Bradshaw, Hyde, & Crowell (2015)12


This investigation employed a web-based survey to enroll 1612 current or former members of the Church of Jesus Christ of Latter-day Saints (LDS) who had engaged in an effort to understand, accept, or change their same-sex attractions. A diverse sample was sought, including participants who reported past engagement in change-oriented intervention. Results indicated that private and religious change venues were far more frequent than therapist-led or group based efforts. Interventions under the auspices of non-mental health professionals were also reported to be the most damaging and least effective. When change of orientation was identified as the goal rather than “understanding” or “accepting” one’s orientation, reported effectiveness was lower.

The authors noted some outcomes (e.g., acceptance of same-sex attraction and reduction in depression and anxiety) that they described as beneficial. Despite that, they said that overall findings supported the conclusion that sexual orientation is highly resistant to explicit change attempts and that SOCE are overwhelmingly either ineffective or damaging. The most ineffective/harmful methods were individual effort, church counseling, and personal righteousness, which consisted of fasting, prayer, and scripture study. The authors concluded their findings are consistent with the APA Task Force’s Report, wherein SOCE is judged as not likely to be effective, SOCE benefit is related to methods not specific to change-related intervention, and therefore only acceptance-based (i.e., gay affirmative) forms of therapy are endorsed.


Researcher bias

Author bias against SOCE is likely since all of the investigators describe themselves as “LGBTQ allies” who “…have been active in supporting the LGBTQ community, online, and national/international engagement.” Four of the five authors were raised LDS, and two remain active in the church. None disclose whether or not they once pursued any form of SOCE themselves. All, however, state that they work closely with LGBTQ Mormons in professional and/or personal roles. This raises the risk that the authors are known by some of the subjects, which increases the likelihood of subject responses in the direction investigators favor. Author bias against SOCE also increases the likelihood of groupthink and the risk of failing to recognize important alternatives, resulting in tainted conclusions and social-policy recommendations.13, 14

Recruitment bias

Dehlin et al. state that they worked with a diverse population sample because they recruited from LDS support groups both in favor of as well as those against psychotherapy for UHA. Since 1992, the Alliance for Therapeutic Choice and Scientific Integrity (formerly NARTH), has been the national professional organization for licensed mental health providers who assist those with UHA. However, the Alliance was not contacted as a source for soliciting participants for this study. Instead, the final sample reflected that 21% of participants were solicited through liberal online and print media (e.g., Huffington Post, Religion, Salt Lake Tribune, and San Francisco Chronicle). Another 21% of the sample was obtained through LDS-affiliated LGBTQ support groups, purportedly across the spectrum of beliefs regarding SOCE. One of those groups, Evergreen International (a group more favorable to psychotherapy) refused to advertise the study, though the authors do not disclose why. Electronic social media and word of mouth led 47% of participants to involvement in the study, which, given the author affinities, cannot be assumed to be representatively divided among opponents of SOCE and those sympathetic to it. Finally, 5% of the sample was solicited through non-religiously affiliated LGBTQ support organizations.

Further sample bias

Additional sample bias is evident in that 71% of participants were either inactive with the LDS Church or separated from it.  This raises concerns about the representativeness of the sample and the response bias this disaffection may have introduced against SOCE. Concerns associated with retrospective, self-report surveys and the fact that 76% were male participants further hamper the reliability and generalizability of results.

Conflated variable scale and midpoint response bias

Another outcome-biasing feature is the manner in which the authors defined their primary outcome measure.  Participants were asked to rate their SOCE experiences on a 5-point scale, where 1 = highly effective, 2 = moderately effective, 3 = not effective, 4 = moderately harmful, and 5 = severely harmful.  This is a highly unusual rating scale in that it is anchored by terms that define different dimensions, i.e., effectiveness and harm. The endpoint outcome measures for a scale are supposed to be opposites (e.g.: effective versus ineffective; harmful versus beneficial). An outcome scale should also include better graduated responses between the endpoints; it is typical to use a seven-point scale. Dehlin and colleagues should have provided participants with two scales. The first should have been anchored by “highly effective” on one end and “highly ineffective” on the other; the second scale should have been bracketed by “significantly beneficial” versus “significantly harmful.”

In addition, the midpoint of a scale is supposed to be neutral, but Dehlin and colleagues’ midpoint is “not effective.” Due to the midpoint response bias, this flawed scale promotes a biasing effect toward SOCE being described as lacking effectiveness.  Midpoint bias refers to the statistical likelihood that respondents tend to choose a middle response when they are pressed for time, uncertain, or lacking an opinion. Seven point scales for both effectiveness and harm that would have allowed for more nuanced responding (e.g., the inclusion of slightly harmful or slightly beneficial, and slightly effective or slightly ineffective options) and neutral midpoints (e.g., neither harmful nor beneficial, and neither effective nor ineffective) would have been more objectively scientific. The conflation of harm and effectiveness in the response scale used in this study creates significant uncertainties about what the results actually mean.

Ideological confounds of Rosenberg’s measure of self-esteem

The authors report they failed to find significant self-esteem differences between participants who had attempted SOCE and those who did not. However, this failure to find a difference may be due to ideological bias inherent in the tool chosen to measure self-esteem. The authors chose to measure psychosexual health in part through Rosenberg’s (1965) measure of self-esteem. Some scales, including Rosenberg’s, define their construct in a manner that is inherently biased against religious values.15, 16, 17 Consequently, scores may reflect differences between humanistic values and theistic beliefs rather than the construct purportedly assessed by the instrument. When the antireligious humanistic dimensions of the Rosenberg scale are statistically controlled, the self-esteem ratings of conservatively religious persons are significantly improved.17 The implication for this study is that self-esteem levels might actually have been higher than indicated for participants who remained conservatively religious. Likewise the “Quality of Life Scale” used leaves out spiritual well-being as a measured quality.

Underrepresentation of professional psychotherapy for UHA

The authors report that religious and private forms of SOCE were far more prominent in their sample than was professional psychotherapy. Whereas 85% of participants indicated engaging in either religious or private individual SOCE methods, only 44% reported some form of therapist or group-led SOCE.  Engaging in “personal righteousness” (such as prayer, fasting, studying scripture, improved relationship with Jesus) was reported twice as much as pursuing professional psychotherapy. Yet the authors report that group-related and therapist-led methods tended to be rated by participants as the more effective and least harmful forms of SOCE.  Furthermore, SOCE “…methods most frequently rated as ‘effective’ tended to be used the least and shortest duration, while methods rated most often as ‘ineffective’ tended to be used most frequently and for the longest duration”.  The authors also contend that this “effectiveness” represented not orientation change but orientation acceptance, decreases in psychological distress, and improvement in family relationships.

The authors used a standard Kinsey scale to evaluate the participants’ sexual orientation. This over-representation of purportedly ineffective/harmful individual (i.e., conducted alone by oneself) and religiously-oriented SOCE methods makes the study’s findings regarding Kinsey ratings and psychosocial health inappropriate as a measure of professionally conducted SOCE.  These general results summed over all SOCE forms therefore are likely to be skewed in an adverse direction, and again might conceal potential positive outcomes of professional SOCE.

Positive outcomes

In spite of the multiple design flaws which bias the study against SOCE, some SOCE methods did receive mildly positive endorsements. Interestingly, these slightly positive ratings were found for therapist-led, group therapy, group retreat, and psychiatric methods. Psychotherapy was found to have moderate or greater effectiveness by 44% of respondents who sought it, with respective effectiveness ratings of 48% for psychiatry, 39% for group therapy, and 48% for group retreats. Of contextual importance is the finding that professional SOCE methods were reported far less frequently by participants than religiously oriented methods, meaning that aggregate results concerning change in Kinsey scores and psychosexual health likely provide an unrealistically negative view of professional SOCE.


This investigation suffers from significant methodological flaws. As a result, it offers no generalizable conclusions regarding psychotherapy for UHA in adults or minors.

SOCE through psychotherapy for LGBQ individuals affiliated with the church of Jesus Christ of Latter-Day Saints by Bradshaw, K., Dehlin, J. P., Crowell, K. A., & Bradshaw, W. S. (2015) 18


No doubt aware of the limitations of the Dehlin et al. (2015) study regarding therapist-led SOCE, this same team of authors analyzed the subsample of respondents who reported participation in psychotherapy for their conflicts regarding same-sex attraction (SSA).  This sample comprised 672 men and 194 women. The authors reported that professional counseling was largely ineffective, with less than 4% of participants reporting any modification of SSA, 42% indicating their change-oriented therapy was “not at all effective,” and 37% finding it to have been moderately to severely harmful.  Homosexuality-affirming psychotherapy was often found to be beneficial in reducing depression, increasing self-esteem, and improving relationships.  The authors conclude that there is a “very low likelihood” of sexual orientation modification and advise highly religious persons with UHA to consider this before pursuing SOCE.


Bradshaw et al. use the same severely flawed dataset employed by Dehlin et al. (2015).  Consequently, the same methodological problems of Dehlin et al.’s original research persist, as well as some additional limitations.

Additional sample bias revealed

Besides the sample bias previously noted, Bradshaw et al. (2015) observed that bisexuality was under-represented in the sample.  This is a concern in that bisexuality is likely to be more responsive to change-oriented intervention than an exclusively homosexual orientation.19 This under-representation could have reduced reports of positive SOCE outcomes in comparison to what might have been obtained with a more representative sample.

Measurement concerns

Outcomes are again measured with the problematic scale that conflates two different dimensions (harm and effectiveness).  The discussion of these concerns noted in the Dehlin et al. (2015) study will not be repeated here. However, their salience can be seen in the authors’ report that 42% of psychotherapy SOCE participants viewed their experience as not at all effective, 21% as moderately harmful, and 16% as severely harmful.  This documentation sounds as if the results are independently derived from two different measures, as they clearly should have been. The fact that they are taken from three neighboring points on a single scale certainly creates the likelihood of a loss of important nuance in the data, thereby unduly inflating participant ratings of harm and ineffectiveness in their evaluations of professional SOCE.  Again, these outcomes surely would have been different had the midpoint been defined as “not at all harmful.” It should also be mentioned that the authors indicate that their survey took, on average, more than an hour to complete.  This fact makes for a greater risk of significant midpoint response bias (which would bias the overall effectiveness rating of SOCE downward) since participants seek to get through an unusually long survey process as quickly as possible.

In addition, Bradshaw et al. trichotomize the goals of psychotherapy-related SOCE into change, acceptance, and understanding.  Yet these are by no means mutually exclusive goals, and it is reasonable to believe that most therapists facilitating SOCE are also promoting goals of acceptance (e.g., of the reality of clients’ SSA) and understanding (e.g., promoting the clients’ self-discovery of the origins of their SSA).  Thus, this forced-choice categorization appears by definition to mischaracterize professional SOCE, again with a likely accompanying loss of data precision that could lend useful refinement to the study’s findings.

Confounding of SOCE forms.

Another serious concern regarding this study is that participants engaged on average in 3.7 non-psychotherapy forms of SOCE interventions which were not differentiated in their overall rating scores. Open-ended responses suggested that some participants applied the outcome ratings narrowly to therapist-led SOCE, while others rated the benefit or harm of their experience across all SOCE forms utilized.  Consequently, the results of this study cannot be reliably linked to professional SOCE, as they may well be adversely distorted by participants’ evaluative inclusion of non-professional and unlicensed providers of SOCE in their ratings.

It is also likely that the 93 participants who reported exposure to an aversive technique in the course of their SOCE experienced this under the direction of unlicensed individuals, or engaged in it years ago when aversive treatments were common to a broad range of clinical concerns within psychology.  Contemporary licensed therapists have long eschewed the use of aversive techniques when assisting those with UHA. This makes it unlikely that the aversive methodologies reported in this survey were facilitated by a licensed mental health provider in recent years.

Additional signs of bias

While not a methodological issue per se, Bradshaw et al.’s discussion of SOCE provides not so subtle indications of their partisan sentiments.  For example, Bradshaw and colleagues dismiss Spitzer’s 2003 research in support of change,20 citing Spitzer’s 2012 “repudiation” of his findings. However, they  fail to note that several of Spitzer’s participants subsequently affirmed their change of orientation and vehemently protested Spitzer’s repudiation of his own 2003 results.21  Bradshaw et al. also cite the demise of Exodus International and admissions of lack of change by its former president.  This is a curious non sequitur in that Exodus was a religious ministry promoting religious forms of SOCE while the present article was supposed to critique only SOCE delivered by licensed mental health providers.  Finally, the authors assert that SOCE requires disregarding the “large body of evidence” that demonstrates “a biological origin for sexual orientation.” Ironically, such a definitive commitment to biological determinism is not even in keeping with the current APA opinion which states, “Many think that nature and nurture both play complex roles….”22


Bradshaw et al. conclude their article with the following statements:

“For adherents to this line of reasoning [i.e., that homosexual attraction may be diminished], the claim of a successful sexual orientation change by a few individuals is sufficient to generalize to the population at large.  The clear evidence, however, is that dutiful long-term psychotherapeutic efforts to change are not successful and carry significant potential for serious harm, and that LGBQ Latter-day Saints find greater satisfaction in counseling approaches that result in acceptance or accommodation.”18

As is evident, the authors first create a straw man argument whereby all SOCE proponents assume that change for some patients means all patients can change. They cite no literature to support this claim but then proceed to challenge this false portrayal by citing the results of their study.  Clearly, this study’s serious methodological weaknesses make the authors’ broad generalizations scientifically unjustifiable.  That Bradshaw et al. would make such unqualified conclusions places their work firmly within the realm of agenda-driven advocacy.


Politics has thwarted the scientific pursuit of quality research on therapy for individuals with unwanted same-sex attraction since the removal of homosexuality from the Diagnostic and Statistical Manual of Mental Disorders in 1973. It is well documented by individuals on both ends of the political spectrum that the American Psychiatric Association’s declassification of homosexuality as a disorder in 1973and by the larger American Psychological Association (APA) in 1975, resulted from the imposition of a sociopolitical agenda, not from the unveiling of new supporting scientific evidence. 23, 24, 25, 26

Nowhere is the APA’s political correctness more clearly displayed than in its “Guidelines for the Prevention of Homophobic Research.” In 1985 the APA’s Committee on Lesbian, Gay and Bisexual Concerns established a Task Force on Non-Homophobic Research which produced detailed guidelines on avoiding research determined to be “heterosexist.” The Task Force defined “heterosexist” as any proposal “conceptualizing human experience in strictly heterosexual terms and consequently ignoring, invalidating, or derogating homosexual behaviors and sexual orientation, and lesbian, gay, and bisexual relationships and lifestyles.”27 The guidelines are prominently displayed on the APA website,28 and its contents are vigorously enforced by the LGB Concerns Committee, whose mission, in part, is “to reduce prejudice, discrimination and violence against lesbian, gay and bisexual people.” As laudable as these social aims may be, it is obvious how such a norm biases the objective pursuit of knowledge regarding all matters related to non-heterosexual attractions and identities. According to Dr. Nicholas Cummings, a past President of the American Psychological Association, the result has been a political correctness that tethers the intellect and a politically correct culture that is more punitive than McCarthyism (p. xv).26

No therapy, whether medical, psychological, or surgical, is 100% effective. All treatments have some degree of failure. In addition, all therapies carry a degree of risk for unwanted side effects. For all forms of psychotherapy used to treat any pediatric mental health concern, there is an estimated 14%-24% deterioration rate among children and adolescents.29 The four investigations reviewed above merely document that some adults experience various “efforts” to change UHA as ineffective and/or harmful. The question to be answered, however, is not “Do some people fail or experience harm?” but rather “Does pursuing the goal of diminishing UHA under the care of a licensed mental health professional result in disproportionate rates of harm and/or failure among minors and adults?” This question has never been scientifically addressed. It is a violation of scientific integrity for the APA, the AAP and others to claim that research proves unacceptable rates of failure or harmful outcomes occur when patients freely choose to diminish unwanted homosexual attractions under the care of a licensed mental health provider.

It is equally outrageous that legislation would be enacted to ban all forms of psychotherapy for UHA with such an absence of scientific evidence or support. Therefore, the College recommends that all such legislation be reversed and that the purview of oversight for non-aversive psychotherapy be left with medical and psychological professionals, and not in the hands of legislators.  The College supports an adolescent’s right to psychotherapy for UHA under the care of licensed mental health professionals. The College, together with the Alliance for Therapeutic Choice, calls for the development of an unbiased research program consisting of investigators from both sides of the sexual orientation debate to ensure that policies promoted by professional medical organizations are rooted in sound science and truly are what’s best for children.

Authors: Christopher H. Rosik, PhD., Michelle A. Cretella, M.D.

January 2016

The American College of Pediatricians is a national medical association of licensed physicians and healthcare professionals who specialize in the care of infants, children, and adolescents. The mission of the College is to enable all children to reach their optimal, physical, and emotional health and well-being.

A PDF of this statement is available here: Psychotherapy for Unwanted Homosexuality Among Youth


[1] Rosik C. The complete lack of a scientific basis for banning SOCE with minors. Aug 2012; accessed 9.17.15 from!analysis-of-anti-soce-legislation/cwgr.

2 American Psychological Association. Guidelines for psychological practice with lesbian, gay, and bisexual clients. American Psychologist Jan 2012;67(1):10-42. Accessed 5/15/15 from

3 Byrd AD. Moving back to science and self-reflection in the debate over sexual orientation change efforts. Social Work 2013;58(1):83-85.

4 American Psychological Association. Ethical principles of psychologists and code of conduct. 2010; Retrieved from

5 Byrd AD, Phelan JE. Facts and myths on early aversion techniques in the treatment of unwanted homosexual attractions. NARTH Institute No year provided. Accessed 12/8/15 from!aversion-techniques-/c4tb .

6 NARTH Institute.  Practice guidelines for the treatment of unwanted same-sex attractions and behavior. Journal of Human Sexuality 2010;2. Oct 2008. Accessed 5/15/15 from

7 Family Policy Institute of Washington accessed 5/15/15 from

8 Haldeman DC. Therapeutic antidotes: helping gay and bisexual men recover from conversion therapies. Journal of Gay and Lesbian Psychotherapy 2002;5(3/4):119–132.

9 American Psychological Association (2009). Report of the task force on appropriate therapeutic responses to sexual orientation. Washington, D.C.  2009; Retrieved from

10  Shidlo A, Schroeder M. Changing sexual orientation:  a consumer’s report. Professional Psychology: Research and Practice 2002;33:249–259. doi: 10.1037/0735-7028.33.3.249.

11 Flentje A, Heck NC, Cochran BN. Sexual reorientation therapy interventions: perspectives of ex-ex-gay individuals. Journal of Gay & Lesbian Mental Health 2013;17:256-277. doi: 10.1080/19359705.2013.773268.

12 Dehlin JP, Galliher RV, Bradshaw WS, Hyde DC, Crowell KA. Sexual orientation change efforts among current or former LDS church members. Journal of Counseling Psychology 2015;1:6.

13 Chambers JR, Schlenker BR, Collisson B. Ideology and prejudice: the role of value conflicts. Psychological Science 2012;24:140-149. doi: 10.1177/0956797612447820.

14  Redding RE. Politicized science. Society 2013;50:439-446. doi: 10.1007/s12115-013-9686-5.

15 Rosik CH.  Ideological concerns in the operationalization of homophobia, Part I: an analysis of Herek’s ATLG-R scale. Journal of Psychology and Theology 2007; 35:132–144.

16 Rosik, CH. Ideological concerns in the operationalization of homophobia, Part II: the need for interpretive sensitivity with conservatively religious persons.  Journal of Psychology and Theology 2007; 35:145–152.

17  Watson PJ, Morris RJ, Hood RW.  Antireligious humanistic values, guilt, and self-esteem. Journal for the Scientific Study of Religion 1987;26:535-546.

18 Bradshaw K, Dehlin JP, Crowell KA, Bradshaw WS. Sexual orientation change efforts through psychotherapy for LGBQ individuals affiliated with the Church of Jesus Christ of Latter-Day Saints. J Sex Marital Therapy Jul-Aug 2015;41(4):391-412.

19 Whitehead NE, Whitehead B. My genes made me do it! A scientific look at sexual orientation. Whitehead Associates. 2010. Retrieved 5/15/15 from

20 Spitzer RL. Can some gay men and lesbians change their sexual orientation? 200 participants reporting a change from homosexual to heterosexual orientation. Archives of Sexual Behavior 2003;32:403-417. doi: 10.1023/A:1025647527010.

21 Armelli JA, Moose EL, Paulk A, Phelan JE. A response to Spitzer’s (2012) reassessment of his 2003 study of reparative therapy of homosexuality.  Archives of Sexual Behavior 2013;41:1335-1336. doi: 10.1007/s10508-012-0032-6.

22 American Psychological Association  (2008) accessed 9/17/15 from

23 Satinover J. Homosexuality and the Politics of Truth. Baker Books, Grand Rapids, Mich. 1996.

24 Bayer R. Homosexuality and American Psychiatry: the Politics of Diagnosis. Princeton University Press. 1987.

25 This American Life on National Public Radio, featuring Alix Spiegle “81 Words” aired 1/18/02; transcript accessed 5/15/15 from

26 Wright R, Cummings NA, eds. Destructive Trends in Mental Health: The Well-Intentioned Path to Harm. Routledge, Taylor & Francis Group. 2005.

27 Herek GM, Kimmer DC. Avoiding hererosexist bias in psychological research. The American Psychologist 1991; 46:957.

28 APA Guidelines for the prevention of homophobic research. accessed 9.17.15 from

29 Lambert MJ. The efficacy and effectiveness of psychotherapy. In Lambert MJ ed., Bergin and Garfield’s Handbook of Psychotherapy and Behavior Change (6th Edition). 2013:169-218. Hoboken, N.J.