Introduction

In part because AIDS has received considerable medical attention, homosexuality is also receiving increasing medical attention. Growing numbers within the health field argue that homosexuality is ‘healthy’ — for individuals and for society. The expansion of this philosophy is exemplified by Dr. John Bancroft’s editorial in the British Medical Journal (1988), “Homosexuality: compatible with full health” in which he decries discrimination against homosexuals, and a National Research Council (NRC) (Turner, Miller, & Moses, 1989) committee’s acceptance and defense of the proposition that homosexuals have shown that “they are not dangers to the society at large” (p. 396). By broaching religious, political, etiological, and mental health issues, Bancroft acknowledged that acceptance or rejection of homosexuality must address considerations of social order as well as public health. In a similarly broad vein, the NRC committee issued sweeping recommendations to the media, churches, educators, and politicians concerning how they should advance societal destigmatization of homosexuality and carriers of the AIDS virus. While Bancroft admitted that “those living a homosexual lifestyle in our society are at greater risk of ill health,” he argued that “[a]part from sexually transmitted disease, this vulnerability is predominantly a consequence of social stigma” rather than part and parcel of homosexuality itself. Similarly, the NRC report decried “animosity toward the sexual behavior of gay men” (p. 396), judged stigmatization of homosexuals a “deeply rooted social pathology” (p. 397), and urged health professionals to become “advocates” of homosexuals (p. 398).

The larger issues alluded to within the Bancroft editorial and NRC report need to be examined. Public health and social order must be factored into any major social decision concerning homosexuality, since these broader concerns, as well as individual health, are often affected by individual choices. Choosing to smoke certainly puts the smoker at ‘greater risk of ill health.’ But it also tends to lower the health of others in his environment and to increase social costs through loss of man-hours and tax revenue associated with earlier mortality. Because both public health and social order are adversely affected by the choice to smoke, society has declared it undesirable, educates against it, and discriminates against those who smoke. Similarly, mandatory seat belt use was upheld by a Nebraska district judge in 1986, partly because society incurs “direct costs of enforcement investigations at the scene of the accident, emergency medical treatment at the scene,” and “indirect costs such as loss of productivity, public welfare, loss of income taxes and rehabilitation.”

Recently, the slowness of the medical community to condemn and discriminate against tobacco smoking has been labelled “a medical, social, and political scandal” (Bailar & Smith, 1986). This brings us to the key issue: is homosexuality compatible with full health (both individual and public) and social order, or might such claims be misinformed? Bancroft’s editorial was challenged as having ignored a host of pathologies highly correlated with homosexual practices (Fowler, 1988; Powell, 1988). We offer new data on the association of homosexuality with the related issues of public health and social order.

Method

In 1983, we performed a one-wave, area-cluster sampling with a questionnaire, investigating 4,340 adults in five metropolitan areas of the USA, chosen for high (Los Angeles, Washington, DC), intermediate (Denver, Louisville), and low (Omaha) levels of homosexual activity as indexed by published homosexual guides to sexual opportunities. One respondent per family unit was asked to complete anonymously an extensive questionnaire on sexuality and return it to the interviewer, sealed in a provided envelope or, if more convenient, to mail it in.

The questionnaire was extensive (over 500 items). The 63 items on which we report data were designed to index: (1) sexual practices (e.g., bestiality, oral/anal contact), (2) sexually transmissible disease (STD) experience (e.g., syphilis, hepatitis) [because homosexuals reported more STD infection for almost every one of the 15 categories (Cameron, Proctor, Coburn, & Forde, 1985), we report only a summary item — “ever claimed an STD”], (3) possible socially disruptive correlates of sexual orientation (e.g., incidence of traffic accidents, shoplifting), and (4) self-abusive/self-harming activities (e.g., tobacco smoking, seat-belt wearing). Almost all of these questions had either been asked or were logically related to questions commonly employed in other questionnaires on sexuality. Data from every item designed to bear upon these issues are reported herein (i.e., there is no selective reporting “to build a good case”).

Two items were combined to index sexual orientation: “How would you rate your sexual desires? not really sexual, not interested/only sexually interested in and attracted to members of the opposite sex (I’m exclusively heterosexual)/generally attracted to members of the opposite sex, but sometimes am sexually attracted to members of my own sex/equally sexually attracted to members of both sexes (I’m bisexual)/generally attracted to members of my sex, but sometimes I’m sexually attracted to members of the opposite sex/only sexually interested in and attracted to members of my own sex (I’m exclusively homosexual)” and “I am a: heterosexual/bisexual/ homosexual.”

Respondents indicating that they were “exclusively heterosexual” on the first item and “heterosexual” on the second were designated as the ‘heterosexual’ group while those who indicated that they were “mainly heterosexual,” “bisexual” or “mainly homosexual” on the first item and “bisexual” on the second item were designated as the ‘bisexual’ group, and those who claimed to be “mainly homosexual” or “exclusively homosexual” on the first item and “homosexual” on the second item were designated as the ‘homosexual’ group. Because our focus was upon possible differences between actively claimed sexual orientations, respondents who claimed to be “asexual” (3% of the total) were excluded from the analysis. Given possible confusion in answering the questionnaire, those who checked “mainly heterosexual” on the first item and “heterosexual” on the second (3% of the total) were also not reported.

The noncompliance/rejection rate was high: 47.5% of the target sample cooperated. Given the nature of the subject matter and the intrusiveness and level of self-disclosure required by the questionnaire, the level of cooperation obtained may approximate the limit of what can currently be achieved in one-wave population surveys regarding personal sexual activity. A similar rejection-rate attended the Fay, Turner, Klassen, and Gagnon (1989) analysis of a National Opinion Research Center [NORC] survey carried out in 1970 (see Note 16). A smaller, multiwave effort using a four-item questionnaire in 1988 achieved a response rate of 72.6% for a nationwide study regarding numbers of sex partners “in the last 12 months,” and apparently even lower rates for other questions about sexual behavior (Centers for Disease Control [CDC], 1988). The CDC sexuality questionnaire was administered only to persons who had already participated in a face-to-face interview concerning nonsexual issues, while ours was introduced as a survey on sexuality. That 5.8% of our male respondents claimed to be homosexual or bisexual while no more than 2% of the NORC one-wave and 3.2% of the CDC multiwave surveys reported homosexual activity in the past 12 months lends support to the belief that we had a sample fairly representative of urban America.

Since every respondent did not answer every item., the Ns vary somewhat from question to question. The median whole age of the three groups was, for heterosexual males 35 years and females 33 years, for bisexual males 32 years and females 28 years, and for homosexual males 32 years and females 29 years. Socioeconomic status as indexed by both educational attainment and wealth was somewhat higher for homosexual practitioners. In the tables that follow, data are reported for all three sexual-orientation groups. However, to test the relationship between sexual orientation and each item of interest, heterosexuals were compared with the combined group of bisexuals and homosexuals [designated ‘homosexual practitioners’] using chi-squared with 1 df. We also report the ‘relative risk’ for appropriate items, defined as the ratio of percentages of homosexuals/bisexuals versus heterosexuals who claimed the experience/behavior of a particular sort (i.e., [bisexuals + homosexuals]/heterosexuals, so for male ‘bondage’ in Table 1, the 25% of bisexuals and 32% of homosexuals combined to 29% divided by the heterosexual 10% and generated a relative risk of 2.9).

Table 1. Claimed Exposure to Individual and Public Health Risks: Percent Ever Reporting Experience

* Groups not significantly different at p < 0.05. For other tested comparisons p < 0.001 by chi-square.

‡ 0 or < 0.5%

Table 2. Claimed Numbers of Sexual Partners and Duration of Sexual Relationships: Median Frequencies and Lengths

Table 3. Indices of Social Disruption/Cohesion: Percent Ever Reporting Experience

‡ 0 or < 0.5%

NS = not statistically significant at p = 0.05

**Estimated Frequencies

Results and Discussion

The relative individual and public health risks reported by each of the three comparison groups of heterosexuals, bisexuals, and homosexuals, separately for male and female respondents, are summarized in Table 1. Table 2 describes the median responses to various items involving numbers of sexual partners and duration of sexually faithful relationships. Table 3 summarizes responses to activities/behaviors that are indices of social disruption or the opposite, social cohesion.

Is Homosexuality “Compatible With All the Criteria of Health”?

(A) If risks to an individual’s health and the health of others are increased the larger the number of different bodies to which one is exposed during sex, then practicing homosexuals more frequently place themselves at greater risk. Comparisons designated ‘Cluster A’ in Table 1 indicate eight activities involving likely sexual contact with persons other than an exclusive sexual partner. Of these 16 comparisons [males and females were counted separately], practicing homosexuals claimed a wider variety of sexual contacts with different bodies in 14, and no statistically significant difference was registered for two activities or behaviors. Practicing homosexuals risked their health more frequently with larger numbers of sexual partners.

While heterosexual males claimed a greater median number of lifetime heterosexual partners than homosexual males, the reverse was true for heterosexual females versus homosexual females (Table 2). Practicing homosexuals claimed a larger total number of sexual partners than the heterosexuals did and reported faithfulness for shorter periods of time. In addition, post-questionnaire inquiry with selected respondents indicated that many homosexuals did not count persons contacted in an orgy or restroom type setting as “partners.” The number of partners claimed by homosexuals may be significantly fewer than the number of bodies with whom these individuals exchanged sexual relations; therefore, it seems likely that our reported differential between homosexual and heterosexual respondents in terms of numbers of claimed partners is an underestimate. Of course, it is biologically irrelevant whether sexual contact occurs within a ‘partner’ or ‘nonpartner’ relationship; number of bodies is the key variable. While no statistical test was applied, these results suggest that homosexual practitioners as a group are a greater risk to public health than avowed heterosexuals, through their sexual contact with a larger number of bodies.

(B) If health risks are inversely related to the cleanliness of the setting chosen for sexual commerce, homosexuality exposes one to greater danger. In six of six comparisons labeled ‘Cluster B’ in Table 1, more homosexuals claimed to have had sex in less sanitary environments in gay baths, peep shows, or public restrooms — than did heterosexuals. The practice of homosexuality was less compatible with public health with respect to cleanliness of environment chosen for sexual activity.

(C) If health risks grow when persons engage in biologically dangerous kinds of sexual activity, that is, sex which involves physical harm to the participants or exposure to bodily fluids and waste produces such as feces, urine, etc., then practice of homosexuality is less healthy. In 34 comparisons listed in Table 1 as ‘Cluster C’ (17 each for men and women), homosexual practitioners more frequently claimed participation in biologically dangerous sex 27 times, five comparisons were not statistically different, and more heterosexuals claimed participation in two activities (male oral/genital sex on female and male anal sex on female).

Participation in an orgy places an individual at considerable biological risk because everyone is apt to be either directly or indirectly exposed to the sexual byproducts of all other bodies in the orgy. In a homosexual/bisexual orgy, feces and urine very often accompany and are spread by the activities. Often feces will be introduced into oral as well as anal cavities when all the participants in an orgy are male (Jay & Young, 1979; Navin, 1981; Suppe, 1981). The biological hazard associated with anonymous sex with others in ‘glory holes’ in restrooms is difficult to estimate but could be substantial since the health status and history of each participant is unknown to the other.

Compounding these health risks is the fact that intercontinental travel and ‘sex tours’ apparently involve a fair number of practicing homosexuals. In an investigation of Danish gays, about a quarter reported sexual contact either with visiting American homosexuals or during visits to the USA during the previous year, and such activity was strongly associated with AIDS seropositivity (Melbye, Bigger, & Ebbesen, 1984). Exposure to feces, semen, etc. from a different society may carry more biological risk than feces, etc. from one’s own, since both the kinds and mix of the various pathogens present might be different from those to which one’s body was adapted. Amoebiasis was once a rare condition in the USA (Quinn, Stamm, & Goodell, 1983), but currently has been reported in about a quarter of homosexual men (McKusick, Horstman, & Coates, 1985), possibly the result of cross-cultural sexual mixing.

(D) Homosexual males more frequently claimed to have ever had a sexually transmissible disease (STD) [the trend for females was in the same direction], and more homosexual females claimed to get high regularly on drugs than heterosexual females [‘Cluster D’ in Table 1].

Is Homosexuality Compatible With Social Order?

If social order is harmed by criminal activities, behavior which endangers lives or which jeopardizes accepted public decorum and disturbs others’ privacy, then homosexuality is associated with social disruption. Of 58 comparisons in Table 3, homosexuals more frequently registered as disruptive 38 times, 19 comparisons among the groups were not statistically significant, and heterosexuals registered as more disruptive once (males for ‘ever in trouble in school’). With few exceptions, in regard to building toward social order and nonendangerment of others, we found differences between heterosexuals and homosexuals favoring heterosexuality. Further, homosexual practitioners exhibited more frequent self-destructive activities.

Sexual relationships with children should also be factored in as bearing upon public health and social order. A review of the literature in English over the past 75 years indicated that about a third of the malefactors implicated in various forms of child sexual molestation practiced homosexuality (Cameron, 1985). In the present sample we found that about 31% of those who reported having been physically sexually molested by an adult while the respondents were under age 13, were homosexually molested (Cameron, Proctor, Coburn, & Forde, 1986). Similarly, in a poorly detailed survey, the Los Angeles Times of August 25-26, 1985 (Timnick, 1985) reported for its nationwide random poll of 2,628 adults that about a third of all molestations were of a homosexual nature. It appears that homosexuals and bisexuals comprise between 3% and 5% of the population of the USA [Cameron, et al. (1985) reviewed the literature concerning the incidence of bisexuality and homosexuality]. Likewise, it appears that about a third of the sexual molestations of children are carried out by practicing homosexuals (Cameron, 1985). Practicing homosexuals then appear to be considerably more apt to involve themselves with children than are heterosexuals, perhaps by a factor of 8 times or more.

While our investigation fell considerably short of the research ideal in many respects, the relationships we describe are similar to those reported in published investigations of volunteers. The first Kinsey study (Kinsey, Pomeroy, Martin, & Gebhard, 1948; Gebhard, Gagnon, Pomeroy, & Christenson, 1965; Gebhard & Johnson, 1979) collected material in the 1940s which suggested that homosexuality is more apt to be associated with various indices of criminality. The Kinsey Institute (Bell & Weinberg, 1978; Bell, Weinberg, & Hammersmith, 1981a, 1981b) and Saghir and Robins (1973) conducted surveys in the early 1970s that yielded a pattern of association between lower public health and higher rates of social disruption and homosexuality. In the only other report of random samples of homosexuals and heterosexuals (Cameron & Ross, 1981 [gathered in the late 1970s]), heterosexuals were less frequently suicidal, less frequently smoked tobacco, less given to substance abuse, less dangerous drivers, and more frequently associated with various indices of social cohesion than were homosexuals. Only a minority of advantages accrued to homosexuals as compared to heterosexuals in any of these investigations. While none of these studies is outstanding from a scientific standpoint, they comprise the totality of comparative studies we could locate; in a word, this is the published database from which comparisons between homosexuality and heterosexuality can be drawn.

Conclusions

From a statistical standpoint, the broad question we investigated was: is sexual orientation independent of the various indices of health and social behavior, or is the alternative true, that homosexual practice is associated with greater health risks and social disruption? Since we compared the groups on so many items, one would expect [at the .05 level] approximately one of every 20 chi-squared tests to be significant even if sexual orientation were unrelated to public health indices. However, of a total of 110 tests, only 32 [those comparisons which were nonsignificant or on which heterosexuals ‘scored more poorly’] did not favor the alternative, so at least 71% of all our comparisons suggest an untoward link between homosexuality and public health and social order.

Is a homosexual lifestyle “compatible with all the criteria of health”? Examining the scientific database, the answer appears to be “no.” Is homosexuality compatible with social order? Our new material fell at least 3 to 1 against homosexuality, and the set of material from the four older studies fell about 4 to 1 against homosexuality. There are undoubtedly advantages to tobacco smoking or drug addiction, but, on balance, so many disadvantages, that society has judged them wrong. The weight of evidence against the homosexual lifestyle appears to be as robust as is that against smoking.

While Bancroft (1988) suggested that those living a homosexual lifestyle were at greater health risk “predominantly [as] a consequence of social stigma” and the NRC committee decried “the particular vulnerability of gay Americans to stigmatization” (Turner, Miller, & Moses, 1989, p. 394), the mechanism by which “social stigma” would produce more frequent bestiality, participation in orgies, hand-balling, or exposure to feces (Table 1) is obscure. How does social stigmatization lead more frequently to having sex to infect others, to larger numbers of sexual contacts, or to more frequent sexual unfaithfulness (Table 2)? How does stigmatization lead to more arrests, shoplifting, income tax cheating, making obscene phone calls, or having sex in public (Table 3)? Almost all of the activities listed in the tables reflect personal choice, and those who so choose bear responsibility.

References

BAILAR JC & SMITH EM (1986) Progress against cancer? New England Journal of Medicine, 319, 1226-1232.

BANCROFT J (1988) Homosexuality: compatible with full health. British Medical Journal, 297, 308-309.

BELL AP & WEINBERG MS (1978) Homosexualities: a study of diversity among men and women. New York: Simon & Schuster.

BELL AP, WEINBERG MS, & HAMMERSMITH MS (1981) Sexual preference. Bloomington, IN:Indiana Univ. Press. (a)

BELL AP, WEINBERG, MS, & HAMMERSMITH, MS (1981) Sexual preference: statistical appendix. Bloomington, IN: Indiana Univ. Press. (b)

CAMERON P (1985) Homosexual molestation of children/sexual interaction of teacher and pupil. Psychological Reports, 57, 1227-1236.

CAMERON P, PROCTOR K, COBURN W, & FORDE N (1985) Sexual orientation and sexually transmitted disease. Nebraska Medical Journal, 70, 292-299.

CAMERON P, PROCTOR K, COBURN W, FORDE N, LARSON H, & CAMERON K (1986) Child molestation and homosexuality. Psychological Reports, 58, 327-337.

CAMERON P & ROSS KP (1981) Social psychological aspects of the Judeo-Christian stance toward homosexuality. Journal of Psychology and Theology, 9, 40-57.

CENTERS FOR DISEASE CONTROL (1988) Number of sex partners and potential risk of sexual exposure to Human Immunodeficiency Virus. Morbidity and Mortality Weekly Report, 37, 565-568.

FAY RE, TURNER CF, KLASSEN AD, & GAGNON JH (1989) Prevalence and patterns of same-gender sexual contact among men. Science, 243, 338-348.

FOWLER AW (1988) Homosexuality. British Medical Journal, 297, 554.

GEBHARD PH, GAGNON JH, POMEROY WB, & CHRISTENSON CV (1965) Sex offenders: an analysis of types. New York: Harper & Row.

GEBHARD PH & JOHNSON AB (1979) The Kinsey data: marginal tabulations of the 1938-1963 interviews conducted by the Institute for Sex Research. New York: Saunders.

JAY K & YOUNG A (1979) The gay report. New York: Summit.

KINSEY AC, POMEROY WB, MARTIN CE, & GEBHARD PH (1948) Sexual behavior in the human male. Philadelphia, PA: Saunders.

McKUSICK L, HORSTMAN W, & COATES TJ (1985) AIDS and sexual behavior reported by gay men in San Francisco. American Journal of Public Health, 75, 493-496.

MELBYE M, BIGGAR RJ, & EBBESEN P (1984) Seroepidemiology of HTLV-III antibody in Danish homosexual men: prevalence, transmission, and disease outcome. British Medical Journal, 289, 573-575.

NAVIN H (1981) Medical and surgical risks in handballing: implications of an inadequate socialization process. Journal of Homosexuality, 6, 67-76.

POWELL, DEB (1988) Homosexuality. British Medical Journal, 297, 555.

QUINN TC, STAMM WE, & GOODELL SF (1983) The polymicrobial origin of intestinal infections in homosexual men. New England Journal of Medicine, 309, 576-582.

SAGHIR MT & ROBBINS E (1973) Male and female homosexuality: a comprehensive investigation. Baltimore, MD: Williams & Wilkins.

SUPPE F (1981) The Bell and Weinberg study: future priorities for research on homosexuality. Journal of Homosexuality, 6, 69-97.

TIMNICK L (1985) The Times poll. Los Angeles Times, 104, August 25-26.

TURNER CF, MILLER HG, & MOSES LE (Eds.) (1989) Sexual behavior and intravenous drug use. Washington, DC: National Academy Press.