Reasons for Caution When Emphasizing Health Disparities for Sexual and Gender Minority Adults in Public Health Campaigns

Many public health campaigns and advocates seek to increase public awareness of well-documented disparities in tobacco use and HIV among lesbian, gay, bisexual, and transgender (LGBT) communities as part of their efforts to reduce these disparities. The National Stakeholder Strategy for Achieving Health Equity advocates this approach. The Centers for Disease Control and Prevention (CDC), the Food and Drug Administration, and state health departments emphasize the magnitude of LGBT disparities in media releases and in some health communication campaigns. In a widely publicized example, national media covered the CDC’s press release emphasizing that “half of black gay men and a quarter of Latino gay men [are] projected to be diagnosed [with HIV] within their lifetime.”¹

POTENTIAL PITFALLS

Although such messages regarding disparities are factually accurate and may raise awareness of a problem, there is burgeoning theoretical and empirical literature to suggest that they may be counterproductive with respect to the larger goal of reducing disparities.^2,3 Campaign messaging that highlights the high prevalence of a behavior can backfire by making the undesired behavior appear normative.⁴ Campaign messaging that highlights poorer outcomes for a minority group may also activate a “stereotype threat” response, thereby reducing performance (e.g., in educational testing) and self-efficacy.³ In the case of LGBT communities, stigma associated with being LGBT contributes to smoking and HIV disparities; messages describing poor outcomes among Black and Latino LGBT populations (as with the earlier-mentioned CDC example) could also exacerbate racial/ethnic stigma.

Few studies have tested how audiences suffering from disproportionate rates of health problems (i.e., those on the negative end of the disparity) process messages emphasizing their disparities.² Overall, such studies show that these audiences tend to have negative emotional responses to such messages and, in turn, reduce their intentions to engage in behaviors that could reduce disparities.^2,5,6 However, this research has been conducted entirely with African American adults and has focused on cancer screening and sexually transmitted diseases. In addition, assessments of health disparity messaging conducted by the Robert Wood Johnson Foundation suggest that talking about health disparities does not resonate well with policymakers.⁷ To our knowledge, no study has examined disparity messaging in LGBT populations.

PRELIMINARY EVIDENCE OF A PROBLEM

We experimentally assessed reactions to three health-related media messages concerning the LGBT community among a sample of LGBT individuals at the North Carolina Pride Festival in Durham in September 2016. A team of 11 data collectors recruited passing festival attendees at a tent on the festival grounds. To be eligible, individuals were required to be aged 18 years or older and to report an LGBT identity.

In total, 225 participants completed the survey (49% male, 75% White, 12% Black, 8% Latino; mean age = 33.4 years; SD = 14.2). We randomly assigned participants to read one of three health-related messages concerning the LGBT community (Table 1) after the prompt “We are now going to ask you to read a message about LGBT communities.” One message was derived from the previously mentioned press release about HIV among gay Black and Latino men,¹ one was taken from the CDC Tips from Former Smokers Web campaign targeting LGBT communities, and we created the third to provide a positively framed contrast about community progress.

TABLE 1—

Mean Reaction Scores With Pairwise Comparisons Between the Three Message Groups: North Carolina, 2016

	Message
Reaction	LGBT Communities Are Working to Address Health Problems in Our Communities, Mean (95% CI)	Half of Black Gay Men Will Get HIV in Their Lifetime, Mean (95% CI)	With Rates Double That of the Population, Smoking Poses a Deadly Threat to the LGBT Community, Mean (95% CI)
The message makes me feel hopeful about the future	3.2 (3.0, 3.4)a	1.2 (1.1, 1.4)b	1.7 (1.5, 2.0)c
The message makes me feel good	3.3 (3.1, 3.5)a	1.1 (1.0, 1.2)b	1.6 (1.4, 1.9)c
The message makes me feel discouraged	1.4 (1.2, 1.6)a	3.1 (2.9, 3.3)b	2.3 (2.0, 2.5)c
The message makes me feel proud of my LGBT identity	3.4 (3.2, 3.6)a	1.7 (1.4, 1.9)b	1.9 (1.7, 2.2)b
I feel insulted by the message	1.4 (1.2, 1.6)a	2.2 (1.9, 2.5)b	1.8 (1.6, 2.1)b
The message makes me feel inspired and encouraged	2.9 (2.7, 3.1)a	1.4 (1.3, 1.6)b	1.8 (1.5, 2.0)b
I am angry about the way LGBT people are described in the message	1.5 (1.3, 1.7)a	2.7 (2.4, 3.0)b	2.2 (1.9, 2.5)c
I believe the message	3.0 (2.7, 3.2)a	2.0 (1.7, 2.2)b	2.2 (2.0, 2.5)b

Note. CI = confidence interval; LGBT = lesbian, gay, bisexual, and transgender. Values within a row with different superscript letters differ at P < .05 according to the Kruskal–Wallis test and pairwise comparison with Bonferroni correction. All items had response options that ranged from 1 (“not at all”) to 5 (“very much”). The sample size was n = 225.

Participants provided reactions to eight ordinal questions used in an earlier study.⁵ Participants self-administered the questionnaire on an iPad via the Qualtrics Offline Survey App with block randomization. Using SPSS version 24 (IBM, Somers, NY), we calculated mean reaction scores for the eight questions among the three message groups and used bootstrapping to calculate corresponding 95% confidence intervals for the mean scores. We conducted eight Kruskal–Wallis tests with pairwise comparisons to examine differences in reactions between the three message groups. Table 1 presents the results, which show poor ratings of disparity messages. For example, the CDC “half of gay Black men” message was rated as not at all believable by 40% of those who viewed it.

Future research on disparity messaging should address the limitations of this pilot study. For instance, generalizability was limited by our use of a convenience sample from a single pride festival. In addition, we tested only three messages, two from actual public health campaigns. Messages were viewed at a single time point in a survey and may not have reflected real-world campaigns. We did not tie our evaluation of measures to a specific behavior and assessed more global emotional and cognitive responses.

EVIDENCE SUGGESTS CAUTION

Our findings, combined with existing evidence,^2,3,5,6 should give pause to professionals considering highlighting disparities. Extensive media coverage was given to a 2016 CDC press release stating that half of gay Black men and a quarter of gay Latino men would be diagnosed with HIV in their lifetime.¹ Among the three messages used in our study, that message in particular prompted undesirable responses, even among our majority White-identified LGBT participants. Although further testing is necessary, our results suggest that messages emphasizing poor outcomes and higher levels of risk behavior (e.g., smoking) may be limited in their effectiveness because they are rejected by the target audience, are simply not believed, or add to the cumulative stress associated with an LGBT identity.

FUTURE DIRECTIONS

Public health professionals should examine additional strategies for using the media to reduce health disparities.² Diverse segments of LGBT communities should also be involved in message development. Although we did not test this possibility in our study, focusing on the normative nature of an undesired behavior may cause it to be perceived as more acceptable.⁴ Future research should examine the impact of normative messages and their potential to backfire, for example, by making smoking seem like something many LGBT people do.

Although also not tested in our study, there is concern that disparity messages may have negative effects beyond the target population. First, the group reported to be at lower risk with respect to a given disparity may feel a reduced level of motivation to change behaviors. Second, policymakers may ascribe the disparity to negative stereotypes, thereby undermining the “call to action” implied in making known a large disparity.⁷ Research should identify effective strategies for talking about LGBT disparities to set an agenda for improvement while minimizing deleterious effects.