Transsexualism : An Unacknowledged Endpoint of Developmental Endocrine Disruption?

Thesis for the Master of Environmental Studies Degree, Christine Johnson, 2004

Johnson is interviewed in Deborah Rudacille’s book The Riddle of Gender. She was born a DES son but transitioned to female (m to f) in her 20s. Scott Kerlin served on her Master’s degree thesis committee at Evergreen State University in Washington and this thesis contained extensive evidence of prenatal exposure to DES, DDT, PCBs, and other endocrine disrupting substances on gender identity and gender development.


In recent years, evidence has accumulated demonstrating that endocrine disrupting chemicals (EDCs) have the potential to alter sexual development at the organizational and functional level in many species, including humans, indicating that this class of chemicals may play a role in the etiology of transsexualism. Although transsexualism has historically been attributed to social or psychological causes, little data exists to support these claims, thus requiring a closer examination of the evidence regarding changes in sexual development due to EDCs. Toward that end, this thesis considers data from studies examining hormonal signaling mechanisms and changes in sexual development observed in wildlife, laboratory animals, and humans exposed to EDCs, all providing a consistent picture that sex hormones and their receptors are highly conserved evolutionarily, finding similar effects of disruption in many species.

Transsexualism: An Unacknowledged Endpoint of
Developmental Endocrine Disruption? by Christine Johnson, A Thesis: Essay of Distinction Submitted in partial fulfillment of the requirements for the degree, Master of Environmental Studies, The Evergreen State College,, June 22 2004.

Image credit Penn State.

In order to place the data in context, a number of historical threads are examined, including: the use of chemicals in agriculture, the use of the pesticide DDT and the pharmaceutical drug diethylstilbestrol (DES), the intertwined relationship between chemical manufacturers and the military, and the history of transsexualism since 1950. The operation and function of the endocrine system is reviewed in order to provide the background to properly interpret findings from endocrine disruptor studies, focusing particularly on the hypothalamic-pituitary-gonadal (HPG) axis. Recent physiological data regarding the vomeronasal organ (VNO) is reviewed, demonstrating that the VNO is the organ responsible for detecting pheromones, or sexually-relevant chemically-based cues, and that exposure of the VNO to extremely low levels of putative sex hormones causes numerous autonomic system responses, including alterations in endocrine function in males. It is therefore suggested that the VNO plays a central role in the circuitry involving sexual development, and a hypothetical framework for testing this concept is provided.

Using this framework, a mechanism for the development of gender identity is proposed, suggesting that gender identity is determined via pheromones by comparing the self with others at an unconscious level. One consequence of this mechanism is that messages conveyed by pheromones can be regarded as signals that can be in contradiction from messages from society, leading to a paradoxical double bind, or a logical contradiction between messages that exist on different logical levels. Another consequence is that there may exist a class of chemicals, pheromone disruptors, that could interfere with pheromones in a manner analogous to endocrine disruptors. Further research must be performed to test this hypothesis since little data exists on pheromones in humans, but early data suggests chemicals may be found that interfere with normal pheromone function.

The prevalence of transsexualism is examined, finding that prevalence differences reported in various countries are not well explained by social factors. Also, it is observed that existing studies have reported the prevalence of transsexuals seeking treatment over a specific time period, but this reporting method is not a measure of the number of transsexuals for each country, which is what the term implies to most people. Several recent epidemiological studies that address sexual changes from endocrine disruption are critiqued, finding that they are plagued with methodological weaknesses and contain a number of errors in interpretation. It is argued that instead of using epidemiological techniques, a more useful approach would be to perform demographic studies that map the birthplace of transsexuals in space and time to determine any patterns that may be related to environmental conditions. The lack of detailed data on transsexual demographics, especially in the United States where such data are completely lacking, has left a void where a lack of data has been interpreted incorrectly as a lack of effect.

The fundamental assumptions used in risk analysis and toxicology are reviewed in the context of recent findings that the effects of a chemical may be larger at low doses than at high doses and that thresholds for the endocrine system must be determined empirically, rather than by assumption of a dose-response curve and extrapolation from an observed toxicological endpoint. The use of invalid techniques by toxicologists has thus invalidated claims of chemical safety, and indicates that public policy based on these techniques are insufficiently protective of public health. Because few things are more important to the continuity of cultures than sexuality and social relations, a number of areas requiring further research are identified, and the need for education of the public is emphasized. I conclude that the existing evidence points towards chemical causes of transsexuality rather than social or psychological causes, requiring a shift in research priorities away from psychosocial studies towards physiological studies of transsexuals.

Hormones, Sexual Development, and Transsexualism

… “this observation raises questions about the validity of existing methods of assessment for studying questions of gender identity development in general. This problem is particularly relevant for offspring of mothers prescribed DES during their pregnancy, since no studies have been performed that specifically asked about changes in gender identity due to these known prenatal exposures to an estrogenic substance.”…

… “The first study was a cohort study of men and women exposed in utero to DES; the study collected demographic data including: age, education, and ethnicity, and psychosexual characteristics including: handedness, sex partner choice, marital status, age at first intercourse, and number of partners.

Interestingly, one finding that reached statistical significance was that DES sons had a higher frequency of left-handedness, which the authors associated with complications during pregnancy; other authors have found that transsexuals also have a higher frequency of left-handedness, suggesting that questions explicitly asking about gender identity status would be appropriate to add to all future studies examining these relations.

However, there are several problems with the study. The format of the study was a survey questionnaire mailed to exposed and unexposed groups, but ultimately, 49% of men and 50% of women from the cohort could not be located, were dead, or chose not to participate in the survey. Consequently, sampling may be biased to an unknown degree because there is no method of determining if those responding are representative of the whole group. Other problems are that all children exposed to DES in utero are assumed to be similarly affected, but because the endocrine disruption thesis suggests that dose and timing is important with regard to outcomes, it is necessary to distinguish when and in what amounts the DES was present in the fetus during critical stages of sexual development. Because the study lacks empirical data regarding prenatal DES exposure, this dramatically weakens the ability of the study to detect differences. Because sexual development proceeds in a number of stages, the timing of the dose is critically important to detecting differences statistically. By considering all those exposed to DES as a homogenous group, effects deriving from exposures during critical periods are obscured, because those exposed outside of these critical windows do not show effects, and therefore they dilute the statistical power of the study. This could be repaired by stratifying the groups by ‘period of exposure’.

Another serious problem with the study is that there is no person in the United States that can be considered a control. As mentioned in the history section, DES was widely used for cattle, thus to a greater or lesser degree, everyone in the U.S. was exposed. Additionally, many other endocrine disrupting chemicals exist, limiting the utility of using a single marker of exposure as an independent variable for changes in sexuality. When these factors are taken into account: high non-response rate, treating all DES exposed persons as a group, and the lack of a proper control, the lack of significance should not be taken as evidence of the lack of an effect.

In a recent teleconference addressing issues of concern to DES sons, Dr. Titus Ernstoff was asked if there were any plans to study gender identity outcomes associated with DES exposure; she replied saying that their study did not have sufficient statistical power to study “rare” outcomes such as transsexuality. However, the DES Sons’ International Network, an internet based research, education, and advocacy group, has found that transsexualism is a common outcome of fetal exposure to DES among sons, indicating that studies on DES exposed persons should be examining endpoints related to altered sexual development. Of their total membership of 600 genetic males, among those who could verify their DES exposure, 52 (8.7%) consider themselves to be transsexual, 26 (4.3%) are transgendered, 9 (1.5%) are gender dysphoric, and 3 (0.5%) are intersex. When all these conditions are considered together, the total rate is 15%, indicating that for DES sons, the frequency of gender identity issues cannot be considered rare. When those who strongly suspect, but who are unable to confirm DES exposure are included, the total number of people who self-identify using one the above categories is 153, or 25% of the network membership.

If the rate of transsexuality in the U.S. is assumed to be 1 in 100,000, and if the number of DES exposed mothers ranges from 3-5 million, the number of transsexuals expected in the entire DES sons population would be 15-25, assuming that DES did not increase the probability of developing transsexualism compared to the general public. Since the DES Sons’ International Network already has double that number of transsexuals alone, this is evidence that among DES sons, the condition is not as rare as Titus-Ernstoff claims, and therefore failure to study the problem may have reasons other than a lack of statistical power – i.e. the decision not to examine the issue may be politically motivated.

Even if a much higher rate of transsexualism in the U.S. is assumed, the numbers do not change very much. For example, if the prevalence of transsexualism in the U.S. is assumed to be the same as in the Netherlands, there should be between 126 and 210 transsexuals in the U.S. due to DES exposure. Because the DES Sons International Network already has 52 transsexuals in their network, this would mean that they have found between 25 and 41% of all transsexuals whose transsexuality can be attributed to DES. Given the membership totals only 600, this seems extremely unlikely, and instead suggests that DES is capable of causing transsexualism, thus explaining the dramatic increase in the frequency of transsexualism in the DES sons group compared to the general public.”

More DES DiEthylStilbestrol Resources

Have your say! Share your views