Emerging Issues in Research on Lesbians’ and Gay Men’s Mental Health: Does Sexual Orientation Really Matter?

American Psychologist

Cochran, Susan D.
IssueNovember 2001
Type of WorkResearch Report


Theoretical writings and research suggest that the onset, course, treatment, and prevention of mental disorders among lesbians and gay men differ in important ways from those of other individuals. Recent improvements in studies of sexual orientation and mental health morbidity have enabled researchers to find some elevated risk for stress-sensitive disorders that is generally attributed to the harmful effects of antihomosexual bias, Lesbians and gay men who seek mental health services must find culturally competent care within systems that may not fully address their concerns.

The affirmative therapies offer a model for intervention, but their efficacy and effectiveness need to be empirically documented. Although methodological obstacles are substantial, failure to consider research questions in this domain overlooks the welfare of individuals who may represent a sizable minority of those accessing mental health services annually.

The relationship between homosexuality and mental health status is a subject that does not want for controversy. In December 2000, the American Psychological Association (APA) published a set of guidelines for psychotherapy with lesbian, gay, bisexual, and other sexual minority1 clients reaffirming the profession's position that homosexuality is not a mental illness

  • (Division 44/Committee on Lesbian,Gay, and Bisexual Concerns Joint Task Force, 2000).

In that same year, articles were published in psychological journals that labeled homosexuality a form of psychopathology (Stone, 2000) or supported the practice of conversion therapy (Nicolosi, Byrd, & Pons, 2000a, 2000b), a therapeutic approach previously condemned by the APA (APA, 1997).

The contradictions embodied by these facts are but a small reflection of the persistent social ambivalence engendered by the topic of homosexuality. A recent
Newsweek poll of the American public found that nearly half of the people surveyed believed homosexuality is a sin (Newsweek Poll, 2000), and approximately a third of those polled in another survey believed it to be a mental or physical illness (Americans on Values, 1999).

In this politicized context, research examining factors related to mental health among lesbians and gay men is extremely vulnerable to biased interpretations (Bailey, 1999).

Yet there are numerous indications within the field of psychology that mental health needs among lesbians and gay men may differ in some important ways from those of heterosexual women and men

  • (Bieschke, McClanahan,Tozer, Grzegorek, & Park, 2000; Cabaj & Stein, 1996; Haldeman, 1994; Hughes, Haas, & Avery, 1997; S. L. Morrow, 2000).

But until quite recently the topic generated surprisingly little of the hard empirical data that might dispassionately clarify psychologists' understandings of mental
health issues affecting lesbians and gay men (Cabaj & Stein, 1996; Rothblum, 1994).

Why Isn't More Known?

There are several obvious and some subtle obstacles to conducting research on the mental health needs of lesbians and gay men.

One is that, historically, the sciences have struggled painfully through changing and often controversial perspectives on homosexuality

  • (American Medical Association House of Delegates, 1996; APA, 1997; Bailey, 1999; Davison, 2001; Garnets, Hancock, Cochran, Goodchilds, & Peplau, 1991; Herek, Kimmel, Amaro, & Melton, 1991; Yarhouse, 1998),

with scientific arguments colored by the strong opinions surrounding the topic that permeate American culture. Homosexuality is widely stigmatized (Herek et al., 1991; Kite & Whitley, 1996), and only 30 years ago it was viewed as a psychiatric disorder reflecting pathological developmental processes (Stein, 1993). In 1974, APA voted to accept a resolution that "depathologized" homosexuality (Conger, 1975).

The controversies may have hindered professional and research development in psychologists' training. In the field of mental health services, clinicians may still be influenced by negative views of homosexuality in their interactions with lesbian and gay male clients and may lack sufficient training in working competently with this population

  • (Bieschke et al., 2000; Crawford, McLeod, Zamboni, & Jordan, 1999; Garnets et al., 1991; S. L. Morrow, 2000).

Researchers, too, who might be interested in studying factors affecting lesbians and gay men are often discouraged from doing so because of both the professional risks that might accrue (Hooker, 1993) and the dearth of available research resources (Solarz, 1999).

A second obstacle is that significant and very real methodological barriers to research with this population exist

  • (Cochran, Keenan, Schober, & Mays, 2000; Solarz, 1999).

Relative to other minority populations, such as racial or ethnic minorities, those with minority sexual orientations are relatively more hidden (Herek, 1998). Consequently, most research has had to rely solely on convenience-based sampling
of individuals who are reachable through their presence in lesbian and gay community venues or through social networks accessible to researchers

  • (Cochran et al., 2000; Cochran & Mays, 2000b).

Often there are no comparable heterosexual groups in these studies, because the methods of sample selection (e.g., recruitment at gay pride events, music festivals,
gay social clubs, gay bookstores, or gay bars) have no obvious counterpart outside the lesbian and gay community.

Sample sizes, too, have generally not been large enough to explore variation in psychiatric morbidity or service use or to examine factors related to this variation.
Research designs that could sample without reference to participation in gay-related social structures are readily available and have been used for years to estimate the mental health of the American population (Manderscheid & Sonnenschein, 1996). But until the onset of the HIV epidemic, which generated public health needs for surveillance of sexual risk behaviors, these periodic population- based surveys did
not directly assess the occurrence of same-gender sexual behavior or sexual orientation identity.

Ostensibly, this was due to concerns about lowering response rates, concerns that
have proved to be unfounded (Butler, 2001). It is only within the past decade that these commonly collected data sets became available for generating estimates of mental health morbidity and services use among lesbian, gay, and bisexual individuals, because questions were added that assessed markers of homosexuality

  • (Cochran et al., 2000; Cochran & Mays, 2000a, 2000b; Cochran, Sullivan, & Mays, 2001; Faulkner & Cranston, 1998; Fergusson, Horwood, & Beautrais, 1999; Garofalo, Wolf, Kessel, Palfrey, & DuRant, 1998; Garofalo, Wolf, Wissow, Woods, & Goodman, 1999; Gilman et al., 2001; Herrell et al., 1999; Lock & Steiner, 1999;Remafedi, French, Story, Resnick, & Blum, 1998; Saewyc, Bearinger, Heinz, Blum, & Resnick, 1998; Sandfort, de Graaf, Bijl, & Schnabel, 2001).

The majority of these recent studies have focused solely on adolescents recruited from middle or high school settings. Even when explicit questions regarding sexual orientation or same-gender sexual experiences are assessed within these general population-based surveys, problems remain.

One is a lack of statistical power due to the low base rate of homosexuality in the population, which, in this context, translates into extremely few lesbians and gay men identified in each study. Population-based studies designed specifically to
compare sexual minorities and heterosexuals on mental health-related
questions are the rare exception (Bloomfield, 1993; Stall & Wiley, 1988).

A second problem is misclassification bias when self-reports of sexual behavior are used as a proxy for sexual orientation. Even the rare misclassification of heterosexuals tends to greatly reduce the positive predictive value of sexual behavior as a screening measure

  • (i.e., the percentage of individuals classified as lesbian or gay by their sexual behavior alone who, in fact, would self-identify as such if they had been asked),

and those who are not sexually active are not classifiable (Cochran et al., 2000).

A third obstacle that has limited psychologists' knowledge of mental health morbidity among lesbians and gay men is, perhaps, an indirect consequence of several studies, beginning in the late 1950s, that repeatedly found few, if any, differences in psychological adjustment between small, nonclinical samples of
homosexual and heterosexual men and women (Hooker, 1993).

This groundbreaking work sidestepped the obvious research bias injected when sampling lesbians and gay men only from psychiatric settings to examine evidence for psychoanalytic writings that emphasized a pathological etiology for homosexuality (Stein, 1993). When findings did not support theoretical predictions of widespread psychopathology among lesbians and gay men, critical empirical support emerged for efforts to remove homosexuality as a psychiatric diagnosis
(Gonsiorek, 1991). But it may also have led to a waning curiosity in research on psychopathology among lesbians and gay men, on the assumption that little of interest would be found.

This is not to say that lesbian and gay men's experiences and needs in therapy were ignored in the published literature. Over the past two decades, mental health professionals have worked extensively to develop specialized treatment modalities, such as gay affirmative therapy

  • (Browning, Reynolds, & Dworkin, 1991; Malyon, 1982),

or theoretical perspectives that focus on the special issues that may arise when lesbians and gay men enter psychotherapy

  • (Brown, 1992; Cabaj & Stein, 1996; Greene & Croom, 2000; Perez, DeBord, & Bieschke, 2000).

Although the work has significantly raised consciousness about issues and experiences of lesbians and gay men, little of this work has had the benefit of empirical study (Bieschke et al., 2000) that could result in evidence-based practice guidelines. [...]