By The National Association for Research and Therapy of Homosexuality (NARTH)
In February 2007, the American Psychological Association (APA) established the Task Force on Appropriate Therapeutic Response to Sexual Orientation (TF). The goal of this six member TF was to answer clinical questions about the efficacy of sexual orientation change efforts. They produced a 138-page document updating and promulgating the APA resolutions of the same title.
The TF deemed that the report was grounded in the scientific fact that same-sex attractions, behavior, and orientations were normal and positive variants of human sexuality. In this view, same-sex orientations do not represent mental or developmental disorders. The TF incorrectly used the research methods of evidence-based medicine to address the following clinical questions: (1) are sexual orientation change efforts (SOCE) effective at changing sexual orientation, (2) are SOCE harmful, (3) are there any additional benefits that can be reasonably attributed to SOCE? The TF broadly defined SOCE and categorized research studies into three designs: experimental, quasi-experimental, and non-experimental. The three categories represented types of quantitative research design. The TF then applied quantitative research design to behavioral research.
The goal of this response is to address concerns about the TF report and the promulgated APA resolutions which the report recommended in an appendix. A major theme of the report, that must not be overlooked, is the driving force of multiculturalism2 or the belief that all cultures are created equal. This ideology allows the APA to assert the null hypothesis as policy; in plain terms, the policy is that homosexuality as culture is no different than heterosexuality as culture. If there is truly no difference between these two "cultures", questions should be asked and answered through scientific inquiry that would allow professionals and laypersons to accept or reject the null hypothesis, i.e. decide rationally whether or not the cultures of homosexuality and heterosexuality differ - and if so how.
Answering these questions scientifically requires adding to the body of knowledge through quantitative or qualitative research. But, in an apparent rush to advance gay civil liberties, the APA ignores these basic questions. However, it is NARTH's position that: (1) basic science relies on quantitative - and sometimes qualitative - data to explain theory and support scientific conclusions. And (2), policy, multiculturalism, or subjective truths cannot be demonstrated, verified or disproved solely by quantitative data alone. The application of the scientific method and the interpretation of its findings should preempt jumping to conclusions where no data exists. This is in keeping with the APA's own "Leona Tyler Principle," which states that in speaking as psychologists, whether as an organization or as individuals, advocacy should be based on scientific data and demonstrable professional experience. Otherwise, psychologists are free to speak individually or as members of a group, but only as "concerned citizens."
Additionally, the importance of preventing biases in scientific research cannot be overlooked. Bias is the over representation or the under presentation of segments of the population. In the postmodern world, this applies not only to the sample, but also to the investigator(s). It is important to note that the TF members consisted of individuals who were on record as opposing re-orientation approaches before being named as a TF member.3 Although a number of APA member psychologists, equally as qualified as those selected, but also experienced in working with those distressed by unwanted homosexual attractions, were recommended to the APA, none were named as a member of the TF. Since the TF included only members with arguably strong gay-activist backgrounds, it may be assumed that this influenced their ability to look objectively at all of the existing scientific data. Although a clear case could be made for confirmation bias simply based on the membership of the TF formally responsible for the report and associated resolutions, the balance of this response addresses the merits of the report itself.
STRONG ASSERTIONS MADE IN THE ABSENCE OF SCIENTIFIC EVIDENCE:
Concerns about the following APA Resolutions
APA Resolution - That the American Psychological Association affirms that same-sex sexual romantic attractions, feeling, and behaviors are normal and positive variations of human sexuality regardless of sexual orientation identity.
APA Resolution - That the American Psychological Association reaffirms its position that homosexuality per se is not a mental disorder and opposes portrayals of sexual minority youths and adults as mentally ill due to their sexual orientation. (APA, 2009, p. 120)
Quantitative research addresses predictions (hypotheses) that are based on the premise that scientific knowledge can be organized into general laws. The TF grounds the aforementioned resolutions in what is claimed as "scientific fact" (dogma/law); namely, that homosexuality is a normal and positive variant of human sexuality (APA, p. 2). Normal is defined by the TF as the absence of mental or developmental disorders. This definition is supported by references from research that shows or claims to show that the homosexual population suffers no more or less mental or developmental disorders than the heterosexual population. The TF makes no mention of more recent and higher quality studies that show that homosexuals do have more mental health issues. When the TF does mention other mental health issues, it deems that these psychological disturbances are caused by "organismic (in-) congruent" religiosity and the stigmatism of a prejudiced society. However, no experimental, quasi-experimental or qualitative data are presented that support their conclusion, let alone even define what are "normal" (typical or usual, if not good, healthy) variations in human sexuality in the overall population. If this type of statistical data exists, the TF should present it in support of their position.4
Additionally, the TF does not define the meaning of a positive variant of human sexuality. They need to specifically define "positive variant", paying particular attention to the positive reproductive advantage of homosexuality. Although it is understood that reproduction is not the only goal of human sexuality, it is likely the most important. The TF should address this oversight before incorporating the word "positive" into formal APA resolutions.
Also missing from the TF's work is a scientific grounding, i.e., suitable references of quality research findings, for the origin of such a variant. For example, if homosexuality is genetic in origin, the human chromosome which contains the specific gene should be identified. It should be stated whether the gene is autosomal or sex-linked. There is no identification of the specific protein product of that gene, and there is no mention of the function of the protein and its influence on behavior. The TF report misses this opportunity to present clearly the scientific, i.e., empirically demonstrable, "facts" or data on which their hypotheses are based.
Though not specifically stated in the report, an implicit hypothesis of the TF is that SOCE have no effect on sexual orientation. This is a correctly stated null hypothesis. However, hypothesis-driven biases are a potential outcome of all quantitative research designs. Functioning from what they believed is a scientific fact that same-sex sexual attractions, behavior, and orientations are normal and positive; the TF's hypothesis would come from this presumably governing scientific law. The TF's "scientific fact" (dogma) established the paradigm that guided the Task Force into their ultimate conclusion not to reject the null hypothesis, i.e, not to accept any evidence of any kind that demonstrated that SOCE may work. With such an initial bias, SOCE's could never be shown or seen to work as their caregivers or recipients intended because, by definition, experience cannot overcome a "scientific fact."5
BIAS IN THE APPLICATION OF EMPIRICAL AND CLINICAL CRITERIA
Concerns about the following APA Resolutions
APA Resolution - That the American Psychological Association concludes that there is insufficient evidence to support the use of psychological interventions to change sexual orientation.
APA Resolution - That the American Psychological Association concludes that the benefits reported by participants in sexual orientation change efforts can be gained through approaches that do not attempt to change sexual orientation.
APA Resolution - That the American Psychological Association concludes that the emerging knowledge on affirmative multiculturally competent treatment provides a foundation for an appropriate evidence-based practice for children, adolescents, and adults who are distressed by or seek to change their sexual orientation. (APA, 2009, p. 120)
Evidence-based medicine research design is a quantitative approach to studying treatment methods involving a cause (independent variable) on some effect (dependent variable). The randomized clinical trial or randomized control trial is the gold standard for sources of new knowledge in evidence-based medicine. The TF deems that their review assessed the current randomized control trials (experimental), nonequivalent group comparisons (quasi-experimental), and multiple uncontrolled designs (non-experimental) as if the methods of evidence-based medicine research were appropriate for evaluating the efficacy or effectiveness of SOCE.6
The TF report includes no SOCE or affirmative-multicultural studies that fit the standards of evidence-based medicine research. No studies are presented that directly compare an SOCE to an affirmative-multicultural therapy. The TF and their report's attempt to evaluate SOCE using evidence-based research in accordance with randomized control design is an inappropriate application of the scientific standard. In light of this, an attempt will be made to address the SOCE studies as presented in the report.
The TF reported that six randomized control trials of acceptable SOCE were completed from 1969 to 1975. In all cases, some form of aversion therapy was the intervention (independent variable) and penile circumference was the effect (dependant variable) measured. Aversion therapy was popular among mental health professionals in the 60's and 70's and was used to treat many types of unwanted behaviors. During that time, there was some use of aversion therapy on persons with distress regarding sexual orientation. However, it was concluded at least 25 years ago that these types of interventions were unethical and did not work in regard to human behavior. In other words, what worked for lower mammals did not work on humans. Yet, the TF seems to imply that these types of therapies are still being used. Also, the TF gives far too much attention to outdated, unethical aversion therapies and too little attention to current approaches to psychological care which are aimed at restoring congruence (attachment). In the TF's view, six randomized control studies of aversion-based SOCE pass the rigor test because of their randomized design. However, these studies are not relevant to gathering data about whether modern approaches to SOCE work. Additionally, the rigorous standard (randomization) set by the TF for SOCE would not be met for research cited by the TF in support of the affirmative-multicultural approaches which it recommends.
In the TF report, affirmative-multicultural therapies should have been presented and subjected to the same standard, presenting their strengths and weaknesses as scrupulously as those of SOCE. This is a significant - and arguably fatal - weakness in the literature review of the report. When used correctly, the methods of evidence-based medicine demand an equitable comparison of one kind of therapy to another therapy. If such a method were relevant for assessing the absolute and relative efficacy/ effectiveness of SOCE and the TF preferred "affirmative-multicultural" therapies, then the TF clearly fails to apply the standards consistently, in an objective or professional manner.
The TF cited three quasi-experimental studies from the years 1971 to 1981. Subjects in these research studies were not randomized, but assigned to nonequivalent treatment groups. These studies did not pass the Task Force's rigor test. Unfortunately, the studies overwhelming relied on aversion therapies as well. These studies only help to show that behavior modification was not the answer. No other answers to the clinical questions can be gleaned from these studies.
The TF further cited thirty-six non-experimental studies from the years 1960 to 1976. In non-experimental design, there is no attempt to control, eliminate, or exclude variables. Again, these studies used the form of behavioral therapy popular at the time, aversion therapy. Many of them were retrospective, i.e., subjects were studied only after they had completed treatment. Overall, as mentioned before, the studies did not support the use of aversion therapy as effective SOCE.
The TF also reported on eight recent studies completed between the years of 1999 to 2004. These studies included various research designs such as retrospective pretest, ethnography, case study, and qualitative retrospective case study. Treatments ranged from conversion therapy to Bible study. Researchers used the patients' perceptions of the usefulness of treatment as a major outcome measure. Many patients reported that they believed sexual reorientation therapy was helpful to them psychologically and physiologically. However, the TF dismissed the relevance of these studies because the design of these studies did not permit cause-and-effect attributions to be made. The TF valued these studies only for their ability to understand the population with distress concerning sexual orientation. To the TF, this population mostly consisted of white men with strong religious backgrounds. Religiosity and stigmatization were the stated motivators of their distress. In the TF's view, this population relied heavily on "telic congruence" (vs. "organismic congruence"); however, the TF does not consider that these studies attempt to investigate the reconciliation of telic and organismic congruence, both of which are belief systems. It is quite possible that developmental and reparative processes including self-awareness and personal identity are not based on either doctrine.
If recent studies using SOCE therapies show that the population of interest can live more congruently with respect to the reality and needs of human physiology and psychology, the dismissal of these older studies is inappropriate. Ultimately, to offer reliable, valid and relevant scientific answers to the clinical questions addressed by the TF, psycho-social and medical clinicians and researchers must study the psychology and physiology of human sexuality as objectively as possible, without undue consideration of religious, societal, family or LGBT values. The TF has not dealt with these issues from an objective scientific stance, but from a belief system based on multiculturalism. In effect, the TF traded science for ideology and activism.
NEGLECT OF CRITICAL AREAS OF SCIENTIFIC LITERATURE
The APA TF report neglected a number of critical areas in the existing clinical and scientific literature. Studies using case study design were ignored in the TF's report. Literature which documents spontaneous change in sexual attraction was omitted. And the literature showing that a combination of factors may be involved in the development of homosexuality was not reported.
Additionally, there was no mention of the literature on the persistence and significantly greater risk of comorbid pathologies in homosexual individuals who live in gay-friendly countries such as Denmark, the Netherlands and New Zealand. Such research casts doubt on the TF's assumption that minority stress is primarily the source of mental-health problems for homosexuals, and suggests that some factors intrinsic to homosexuality may instead be at work in elevating the level of mental-health problems. There also was a neglect to discuss the literature on lack of relationship commitment and the relational instability among homosexual individuals, including among those who have been legally "married" or otherwise given formal civil recognition in a same-sex union.
The TF failed to discuss predictive factors in the development of homosexuality and the hypothesized mechanisms of change in SOCE interventions. No attempts were made to reconcile the APA Resolutions to new theories on the development and regulation of the central nervous system or the origin of self. Scientific knowledge has significantly increased in these areas, and the TFs should have considered them. Literature on the etiology of homosexuality (e.g., the influence of family interactions) was disregarded as if it had been scientifically disproved. The TF report also arbitrarily disregards a large subset of the literature on SOCE, presumably due to its age. Such older research was state of the art at the time and warrants that this literature be considered more seriously.
SUMMARY
Basic scientific research in all human development has advanced tremendously and is ongoing. Modern medicine owes its existence to the quantitative design of basic scientific research. As professionals, we cannot ignore this knowledge. If real theories are to emerge on the true origins of a variety of human behaviors and experiences, clinicians and researchers will have to reconcile this massive increase in knowledge. It is questionable whether a postmodern society can do this type of work.
Perhaps a future endeavor of the APA should be to design and conduct an evidence-based research study which compares a relevant SOCE to a relevant affirmative-multicultural therapeutic approach. The APA insists that affirmative multicultural therapies can provide the same relief as SOCE; however, this only can be scientifically evaluated through an evidence-based research project. Such a project ideally would use the methodological rigor emphasized by the TF. In this way, the patients would be randomized to groups, SOCE could be the intervention, affirmative multicultural therapy would be the comparison, and objective and subjective outcomes would determine the relative efficacies between the therapies. This type of study would add to the body of knowledge needed in helping to answer the important clinical questions on SOCE.
It also should be noted, however, that a true experimental test of the absolute and relative efficacy/effectiveness in the end may be therapeutically unethical to conduct. The rights of clients to self-determination, i.e., to decide their own goals of treatment, may not be respected if clients were randomly assigned to be treated for a goal they did not want. For example, it would be unethical to force persons who do not want relief from homosexual attractions and/or behaviors to undergo therapy with that goal in mind. Similarly, it would be unethical to force a client who wanted psychological care to resolve unwanted homosexual attractions/behaviors to undergo instead therapy to enable them to accept and be pleased with those attractions/behaviors.
The conclusions of the APA Task Force are based on a postmodern belief in multiculturalism, in which traditional science is looked at with skepticism and "truth" is in the eye of the beholder. The poor use of science in the TF report appears to be yet another example of a disturbing trend. As Baker, McFall and Shoham (2009) have argued, "Clinical psychologists' failure to achieve a more significant impact on clinical and public health may be traced to their deep ambivalence about the role of science and their lack of adequate science training, which leads them to value personal clinical experience over research evidence" (p. 8). In his introductory commentary to the Baker, et al., monograph, Walter Mischel laments that this "widening gulf" (p. 1) or "disconnect between much of clinical practice and the advances in psychological science is an unconscionable embarrassment" (p. 2). Mischel further warns that "clinical psychology... will increasingly discredit and marginalize itself if it continues the trajectory it has pursued for far too many years" (p. 1).
REFERENCES:APA Task Force on Appropriate Therapeutic Responses to Sexual Orientation. (2009). Report of the Task Force on Appropriate Therapeutic Responses to Sexual Orientation. Washington, DC: American Psychological Association.
Baker, T.B., McFall, R.M., & Shoham, V. (2009). Current status and future prospects of clinical psychology: Toward a scientifically principled approach to mental and behavioral health care. Psychological Science in the Public Interest, 9 (2), 5-145. Retrieved on 04/21/10 from: http:// www.psychologicalscience.org /journals/pspi/inpress/baker.pdf.
Mischel, W. (2009). Editorial: Connecting clinical practice to scientific progress. Psychological Science in the Public Interest, 9 (2), 1-4. Retrieved on 04/21/10 from: http:// www.psychologicalscience.org /journals/pspi/inpress/baker.pdf.
Tyler, L. (1969).An approach to public affairs: Report of the Ad Hoc Committee on Public Affairs. American Psychologist, 24(1), 1-4. doi: 10.1037/h0037787
ENDNOTES:(1) This represents a formal, overarching scientific response by NARTH to the APA's Task Force report. NARTH already has responded with a shorter statement: http://www.narth.com/docs/apataskforcereportbroch.pdf. This paper is not exhaustive, Other commentaries on particular aspects of the APA report by individual NARTH members may be seen at: http://www.narth.com.
(2) Multiculturalism is part of a postmodern ideology or worldview (Zeitgeist) in which traditional scientific inquiry is devalued and replaced with subjective "truth".
(3) Judith M. Glassgold sits on the board of the Journal of Gay and Lesbian Psychotherapy and is the past president of APA's Gay and Lesbian Division 44; Jack Dresher is a well-known gay-activist psychiatrist; Lee Beckstead is on record as opposing any efforts to change sexual orientation and is a gay-identified man; Beverly Green was the co-editor of APA Gay and Lesbian Division 44 series on Psychological Perspectives on Lesbian, Gay, and Bisexual Issues; Robin Lin Miller worked for the Gay Men's Health Crisis and has written for gay publications; and Roger Worthington is the interim Chief Diversity Officer at the University of Missouri and was awarded the "2001 Catalyst Award" from the LGBT Resource Center.
(4) In more technical terms, what is lacking is a predictive population frequency of variations in human sexuality that could be defined as normal (no reference range). No descriptive statistics are presented that define a Gaussian distribution or probability of population statistics relative to human sexuality. As presented, the definition of "normal variant" could be misinterpreted that homosexuality falls within two standard deviations of a human sexuality distribution; however, it is much more likely that homosexual behavior falls into or near the tails of a normal distribution. For example, for a trait that was "normally distributed" (e.g., height), most or roughly 68% of persons would be within one standard deviation of the "mean" (average) height common for someone of that sex, and "almost all" or roughly 95% of persons would be within two standard deviations of their mean height. Statistically, the TF lacks the research to claim that homosexuality is "normal", i.e., statistically "not uncommon".
(5) The analysis in this section is also relevant for considering the implications of the TF report's claim of a second "scientific fact": "Gay men, lesbians, and bisexual individuals form stable, committed relationships and families that are equivalent to heterosexual relationships and families in essential respects" (APA, p. 2). Cited references, for this and the other "scientific fact" mentioned in the report, include only the political decision to remove "homosexuality" from the DSM-II, APA resolutions, and opinion pieces by gay activists as references. None of these, singly or taken together offer sufficient proof for these TF assertions.
(6) In general, it is methodologically difficult to prove a cause and effect relationship between variables that are meant to impact human behavior. Experimental (quantitative) data link independent variables to dependent variables (in this instance, SOCE to change in one or more facets of sexual orientation). This makes it difficult to show if and the extent to which SOCE may impact sexual orientation. This is especially true in the light of the experimental data that is presented in the report. The TF lumps the outcome research findings of all the SOCEs which are reported together. Also, the TF excludes other clinical and research data which otherwise support the efficacy of SOCE. Methodologically, the TF has applied reasoning that sets the stage for a Type II error, i.e., failing to notice significant or meaningful change when it does occur. Again, quantitative research design is difficult to use in the study of any behavior, including sexual orientation.