HIV-Related Stigma in Midlife and Older Women

Abstract

HIV-related stigma has been identified as a barrier to HIV testing and prevention efforts internationally and nationally. Although the prevalence of HIV/AIDS is rapidly increasing in women aged 50 and older, little is known about the HIV-stigmatizing behaviors in this underserved population. In this exploratory analysis we investigated the influence of self esteem, sensation seeking, self silencing, and sexual assertiveness on the HIV-stigmatizing behaviors in 572 women aged 50 to 93. The variables considered in this analysis included subscales for the variables self silencing (Externalized Self Perception, Care as Self Sacrifice, Silencing the Self, Divided Self) and sexual assertiveness (Information Communication, Initiation, Refusal). Contributing predictive variables in the final model included: Self esteem (β = −.190, t = −4.487, p = .000), Care as Self-Sacrifice (β = .125, t = 2.714, p = .007), and Silencing the Self (β = .164, t = 3.290, p = .001). Initiation (β = .108, t = −2.619, p = .009) and Refusal (β = −.091, t = −2.154, p = .032). Implications for social work and health perception regarding HIV prevention in midlife and older women are considered. Development of age- and gender-appropriate strategies assisting women in addressing HIV-stigma and related influences are addressed.

KEYWORDS:

• HIV
• stigma
• aging
• midlife
• older
• women
• HIV risk factors
• sexual assertiveness
• self silencing

New estimates from the Centers for Disease Control and Prevention (CDC) indicate that approximately 56,300 new HIV infections occurred in the United States in 2006 (CDC, 2008). Eleven percent of all reported new HIV infections occur among women aged 50 and older (CDC, 2008). The majority of older women, like younger women, acquire the virus through heterosexual contact (CDC, 2008). More than half of women infected who are midlife and older are of African-American and Hispanic origin, indicating greater risks among minority groups (CDC, 2007).

However, many older women believe that HIV has little personal relevance, possess limited knowledge of age and gender specific risk factors, and profess HIV-related stigma (Hillman, 2007). HIV-stigma has been associated with inconsistent condom use in midlife and older adults (Illa et al., 2008). HIV-stigma acts as a barrier to testing and to preventive strategies. Older adults at high risk for HIV infection are less likely to engage in HIV prevention strategies than younger individuals and are less likely to be tested than their younger counterparts; this difference has been thought to be due to the stigma attached to HIV (Emlet & Farkas, 2002; Mack & Ory, 2003). However, few studies to date have provided information about HIV stigma in midlife and older women.

MIDLIFE AND OLDER WOMEN'S SEXUALITY

The results from the National Council on Aging (2000) indicated that sexual activity plays an important role in relationships among older adults. Many women aged 50 and older are sexually active (Lindau, Leitsch, Lundberg, & Jerome, 2006; Sormanti & Shibusawa, 2007) and have the capacity to remain so as they age (Gott & Hinchliff, 2003; Meuris, 2007; Montenegro, 2003; Robinson & Molzahm, 2007). While sexual activity decreases with age, older women regard sexuality as an important part of life (Lindau et al. 2007). When older women are not sexually active, it is usually because they lack a partner or because they have a limiting medical condition (National Council on Aging, 2000). Advances in postmenopausal treatment and other medical interventions have enhanced sex and sexuality of midlife and older women, including the use of erectile dysfunction medications for men (i.e., sildenafil). Due to freedom from worry about unintended pregnancy, postmenopausal women may not use contraception (i.e., condoms) during intercourse (Emlet & Farkas, 2001; Lindau et al., 2006; Zablotsky & Kennedy, 2003), increasing the risk of infection. Compared to younger persons, adults aged 50 and older are only one-sixth as likely to use condoms (Poirot, 2002).

The expectations and behaviors of older women are likely to change over time. While previous cohorts might be of the opinion that sex was something improper and not to be discussed (Cranston & Thin, 1998), baby boomers (i.e., those born between 1946–1962) who reached sexual maturity during the sexual revolution of the 1960s and are comprised of women now in their late 40s and 50s may not. This group of women, who were much more sexually explorative in their youth and succeeded in detaching sexuality from reproduction, may be more comfortable with sex and sexuality in general. However, they may be more apt to engage in risky sexual behaviors due to their comfort and reduced inhibitions.

HIV Risk in Midlife and Older Women

An ever-increasing number of studies have identified various personal and psychosocial factors that may influence HIV risk and protection in women, but few have directly addressed the special needs and considerations of midlife and older women. Additionally, several female-specific biological characteristics may enhance the efficiency of heterosexual HIV transmission in older women, such as physical changes that occur in menopause (i.e., drying and thinning of the vaginal walls that may cause abrasions and tears that increase the risk of infection through unprotected sex) and the presence of a sexually transmitted disease (STD) (World Health Organization, 2003). Moreover, among HIV-discordant couples, heterosexual transmission from male-to-female is approximately eight times more efficient than female-to-male transmission (Padian, Shiboski, Glass, & Vittinghoff, 1997).

Many older women do not have sufficient knowledge regarding HIV transmission and therefore may not perceive themselves at risk for infection (Savasta, 2004). This perception of invulnerability can influence sexual decision-making and undermine HIV protective behaviors (Ory, Zablotsky, & Crystal, 1998; Williams, Ekundayo, Udezulu, & Omishakin, 2003). Williams et al. (2003) found that 98% of the 745 women participants aged 55 and older perceived their chances of becoming infected from low to zero probability.

Some women have become infected through sexual contact with bisexual men who may not disclose their high-risk sexual behaviors (Hader, Smith, Moore, & Holmberg, 2001; Wohl et al., 2002; Wolitski, Jones, Wasserman, & Smith, 2006). The older the woman, the less she perceives herself at risk for infection. Midlife and older women are less likely than their younger counterparts to understand HIV risk behavior and to be aware of their partners' HIV status (Binson, Pollack, & Catania, 1997). The factors that were investigated in this study include HIV-stigma, self esteem, sensation seeking, self silencing, and sexual assertiveness.

Factors That Influence Risk and Protection in Midlife and Older Women

HIV-Stigma

HIV-stigma has been defined as “prejudice, discounting, discrediting and discrimination directed to people perceived to have AIDS or HIV, and the individuals, groups, and communities with which they are associated” (Herek, 1999, p. 1107). Stigmatizing is typically enacted against a person or group out of prejudice and it functions in relationships to diminish the person or group being targeted. The social notion of deviance—or merely being different from the majority— signifies that the stigmatized person is devalued (Goffman, 1963).

Stigma is an interactional process that exists within a social or cultural context. There is a perception that the person is responsible through personal choices for acquiring the condition. The un-stigmatized believe themselves to be superior to the people who are targeted. Stigmatizers blame individuals from groups that are traditionally associated with HIV (i.e., gay men, drug users, sex workers) for their behavior and in doing so may feel reassured about their own invulnerability to disease (Parker & Aggleton, 2003).

HIV is associated with a high level of stigmatization (Herek, 1999) whereby individuals are often blamed for their affliction. People, not properly educated on its transmission, may feel threatened by HIV and stigmatize those with it. In the case of HIV, stigma and discrimination can lead to fear of disclosure, inadequate and inappropriate treatment, perception of low risk demonstrated by distancing oneself from “high-risk groups” and further spread of a disease which might otherwise be contained.

Stigma associated with HIV may be particularly problematic for midlife and older women. They may be reluctant to discuss HIV and high-risk sexual behaviors because of the stigma society attaches to the disease (Montoya & Whitsett, 2003). HIV-related stigma also bolsters age-old fears and prejudices directed at those with life-threatening conditions and those who flout society's rules. The dread of contracting a fatal illness is thus intertwined with the shame that surrounds a condition whose modes of transmission include sexual contact and drug use—two factors subject to judgments about social norms, including appropriate sexual behavior.

An increasing number of single, older American women are maintaining active sex lives because of healthier lifestyles and erectile dysfunction treatments. Women who do not use protection put themselves at risk for HIV. There is a stigma stemming from embarrassment of having gray hair and still being sexually active. These stigmas act as barriers for older women, since they face a double stigma of ageism and the idea that older people should not be having sex.

HIV-stigma can interfere with effective prevention efforts and individual behavior change. Although empirical research in this area is scant, it appears that fears related to HIV-stigma can dissuade individuals at risk for HIV from being tested and for seeking information and assistance for risk reduction (Herek et al., 1998). Delays in testing increase the likelihood that infected persons will transmit the virus to their sexual partners. Internalized stigma and fear of stigma from sexual partners with whom they broach the subject of safer sex practices may also increase the likelihood of continuing sex behaviors that transmit HIV. These findings support the idea that stigma is related to greater behavioral risk for HIV acquisition. Stigma may cause older women in particular to ignore or distance themselves from a disease for which they may be at risk.

There is a paucity of published research on older women and HIV and even less on HIV-related stigma. This gap is due to the focus of researchers on groups highest at risk for HIV (Emlet, 2005). In the few studies that have included older adults in research to measure HIV-stigma, the focus has been on those persons already living with the virus (Emlet, 2005; Heckman et al., 2002). HIV-stigma is a psychosocial phenomenon that is experienced differently from person to person, and it is unclear what influences HIV-stigmatizing behaviors.

Self esteem

Many women experience a decline in self esteem in later life (Gaylord, 2001). Older women experience a more negative self concept as well as lower life satisfaction, happiness, and self esteem compared with older men (Pinquart & Sorensen, 2001). Midlife and older women may face communication barriers about safer sex based on the perceptions of their loss of “currency” in dating and sexual relations (e.g., losses due to ageism, poor body image, or diminished sense of social power in contrast to male partners) (Coon et al., 2003). Moreover, older women are more likely to be widowed than older men (Pinquart & Sorensen, 2001), placing them at risk for diminished self esteem following the critical loss of a spouse (Carr et al., 2000; Fry, 2001; Lund, Caseta, & Dimond, 1993; Quandt, McDonald, Arcury, Bell, & Vitolis, 2000; van Baarsen, 2002). Loss of meaningful spousal role and associations with friends of the deceased, diminished economic resources, and interpersonal dependency of the widowed spouse may increase vulnerability and lead to decline in self esteem. Few studies have investigated the influence of self esteem in older women with regard to their HIV attitudes and risk-taking behavior (Jacobs & Thomlison, 2009).

Sensation seeking

Zuckerman (1994) defined sensation seeking as “a trait defined by the seeking of varied, novel, complex, and intense sensations and experiences, and the willingness to take physical, social risks, legal, and financial risks for the sake of such experience” (Zuckerman, 1994, p. 27). Sensation seeking has been identified as an important correlate of sexual risk behaviors (Bancroft et al., 2004; Crawford et al., 2003; Donohew et al., 2000; Gonzalez et al., 2005; Gullette & Lyons, 2005; Hoyle, Fejfar, & Miller, 2000; Kalichman, Cain, Zweben, & Swain, 2003; Kalichman, Heckman, & Kelly, 1996; Rolison & Scherman, 2003; Thompson, Kao, & Thomas, 2005) yet there is little information on gender differences in these relationships. Although, on average, men tend to be higher in sensation seeking than women (Bell, O'Neal, Feng, & Schoenrock, 1999; Gullette & Lyons, 2005; Zuckerman, 1994), many women are high sensation seekers and an increasing number of women take health risks such as binge drinking (Pirkle & Richter, 2006; Wechsler et al., 2002). There are virtually no studies that have studied age differences in sensation seeking behaviors. Zuckerman, Eysenck, and Eysenck (1978) found that sensation seeking declines in women (Zuckerman, 1994). However, this study was conducted with women who came into adulthood before the feminist movement; the decline may be socio-generational rather than age-related. It is important to note that the youngest women in this current cohort would have been born in the late 1950s and would have been in their 20s during the height of the sexual liberation and feminist movements of the 1970s. Because few studies have examined age differences in sensation seeking behaviors, sensation seeking was taken into consideration when trying to predict factors that influence HIV-stigma in midlife and older women.

Self silencing

Self silencing is how women silence or suppress thoughts, feelings, and actions that clash with those of their partner in order to avoid conflict and maintain a relationship (Jack, 1991; Jack & Dill, 1992). According to Jack (1991), self silencing is due to the belief in certain women that the needs of others are more important than their own. This maladaptive behavior renders a woman outwardly compliant and may impede her ability to negotiate safer sex practices with her partner, increasing her risk for HIV infection.

The four categories of socially constructed self silencing behaviors proposed by Jack and Dill (1992) are: (a) Externalized Perception of Self (i.e., judging oneself by external standards), (b) Care as Self Sacrifice (i.e., placing the perceived needs of others before oneself in order to preserve the relationship), (c) Silencing the Self (i.e., inhibiting one's self expression by governing personal behavior in intimate relationships), and (d) The Divided Self (i.e., maintaining an outer compliant self to conform to social norms about how a woman should act while she is experiencing anger and hostility). According to the concept of self silencing, some women believe that self expression is unacceptable to their partners, so in order to maintain these relationships they develop the self silencing behaviors addressed by the four schemas. Self silencing may be linked to social values and cultural norms that encourage women to adhere to traditional gender roles (Witte & Sherman, 2002). Older women may have particular difficulty in communicating about sexual behavior in general (Zablotsky & Kennedy, 2003) and STDs such as HIV. This lack of dialogue between older women and their partners creates a barrier to negotiating safer sex practice and inhibits discussions about HIV. Silencing behaviors and transitions to action have not been studied in the growing population of older women at risk for HIV.

Sexual assertiveness

Sexual assertiveness is a direct expression related to communication strategies that protect an individual's sexual health by claiming physical autonomy. It is rooted in the idea that women have rights over their bodies and freedom over sexual expression (Ricket, Sanghvi, & Wieman, 2002). If a woman does not believe she has the right to assert her sexual rights and protect herself from HIV infection, her risk increases. Sexual rights include whether or not a woman wishes to have sex, the type of sex she engages in (e.g., protected sex using condoms), and with whom she desires to have sexual relations. Perceived or actual lack of power in sexual relationships is likely to negatively affect safer sex negotiation. Sexual assertiveness has been shown to play an important role in safer sex decision making and behaviors. Women with higher sexual assertiveness are less likely to have engaged in unprotected vaginal sex (Ehrhardt et al. 2002; Noar, Morokoff, & Harlow, 2004; Pulerwitz, Amaro, De Jong, Gortmaker, & Rudd, 2002). Midlife and older women living with HIV have reported they did not know the HIV-risk histories of their sexual partners, including whether they had been tested or not (Neundorfer, Harris, Britton, & Lynch, 2005). These data imply that negotiating condom use or discussing HIV risk practices—both practices with the potential to require some degree of sexual assertiveness—may be especially difficult for older women who grew up in an era when sexuality was not openly discussed.

For midlife and older women in long-term, monogamous relationships, communicating assertively about sexual issues may still be problematic. Most committed heterosexual couples, regardless of age, do not feel the need to use condoms (Morokoff et al., 1997) or find they interfere with male sexual functioning (Savasta, 2004). The research on sexual assertiveness and older women, albeit scant, suggests that women in this age group are less likely than their younger counterparts to be sexually assertive and may lack the ability to successfully negotiate safer sex practices (e.g., condom use) (Neundorfer et al., 2005).

Older adults in general have been overlooked in terms of HIV prevention as efforts have been targeted to younger populations (Linsk, 2000). Because there is paucity in these areas, the goal of the study was to more clearly assess the nature of relationships among HIV-stigma and relevant psychosocial variables; that is, to explore how stigma is informed and impacted by self esteem, self silencing, sensation seeking, and sexual assertiveness for the purpose of HIV prevention in women aged 50 and older.

METHODS

A cross-sectional, correlational research design was used in this study using multiple regression to explore the influence of selected psychosocial variables (self esteem, sensation seeking, self silencing, and sexual assertiveness) on HIV-related stigma in midlife and older women.

Data Collection Procedures

Primary data were collected from September 2005 through February 2006 by anonymous written questionnaires using a community based sample of midlife and older women (aged 50 to 93) residing in South Florida. The setting for this study was a variety of community sites where midlife and older women frequently visit (e.g., women's clubs, hair salons, health clinics). Potential participants were approached face-to-face by the researcher or assistant. Once the women agreed to participate in the study, they were each given a questionnaire, a pen or pencil, and an envelope. Participants were given instructions verbally and in writing. These instructions communicated the purpose of the study, consent to participate, benefits of and risks related to participation, directions for completing the questionnaire, an assurance of anonymity, and plans for the dissemination of research findings. The researcher and trained research assistants gave each woman an envelope in which to place the completed questionnaire. The questionnaire was available in English or Spanish. The women were not tested for HIV.

The participants were assured that no personal identifiers (e.g., names, address, and telephone or e-mail contacts) would be requested and were instructed not to put their names or other personal identifiers anywhere on the questionnaire or envelope. During data collection, the researcher or assistant were available to answer questions. Participants were instructed to place the completed questionnaire in the envelope, seal it, and drop it in a questionnaire collection box supplied by the researcher (or trained research assistant). Participants received a small gift in appreciation of their time.

Instrumentation

The predictor and dependent variables used in this study were operationalized and measured using several validated instruments. A brief description of each measure is provided.

Rosenberg self esteem scale (RSES)

With 10 items using a four-point Likert-type response set, this instrument measures levels of self esteem. The scale has demonstrated high internal consistency with Cronbach's alpha ranging from .67 to .94 and high test–retest reliability of .85 for three month periods. Higher scores indicate greater levels of self esteem. For detailed information about this scale, consult Rosenberg (1965). In this sample the mean score was 33.0 (Min = 2, Max. = 40, SD = 4.87). The women in this study had relatively high levels of self esteem.

Sensation seeking scale (SSS)

This scale consisted of 13 items and used a dichotomous forced-choice format. With a possible range from 0 to 13, higher scores represent higher sensation seeking. Reported internal consistency for this instrument was found to be .85 (Zuckerman, 1978; Zuckerman et al., 1978). In this study, the mean score was 4.1 (SD = 2.77, Min. = 0, Max. = 13), indicating relatively low levels of sensation seeking in this sample of older women.

Silencing the self scale (STSS)

This scale was developed by Jack and Dill (1992) and measures self silencing behaviors. It consists of 31 items divided into 4 subscales. Each item used a five-point Likert-type scale (1= strongly disagree; 5 = strongly agree). With a possible range from 31 to 155, higher scores indicate greater silencing behaviors. Jack and Dill (1992) report this instrument to possess test–retest reliability, internal consistency, and construct validity. In this sample, total scores ranged from 31 to 137 (M = 78.6, SD = 20.68). Subscale scores include Externalized Self Perception (Min. = 6, Max. = 30, M = 14.6, SD =5.62); Care as Self Sacrifice (Min. = 9, Max. = 43, M = 24.9, SD = 6.09; Silencing the Self (Min. = 9, Max. = 42, (M = 22.8, SD = 7.58); and Divided Self (Min. = 7, Max. = 35, M = 16.4, SD = 6.07). The Externalized Self Perception subscale addresses the tendency to self-evaluate using external standards. The Care as Self-Sacrifice subscale incorporates items that measure the tendency to put the needs of others before one's own needs. Items that measure the tendency to avoid conflict in order to preserve a relationship are addressed in the Silencing the Self subscale while items that address internal turmoil while maintaining observable compliance are considered in the Divided Self subscale.

Sexual assertiveness scale (SAS)

This instrument was developed by Morokoff et al. (1997) and measures sexual assertiveness. It is comprised of 18 items and is reported to possess good internal consistency, test-retest reliability, and construct validity (Morokoff et al., 1997). It is divided into three subscales. Items are scored using a five-point Likert-type scale (1 = never; 5 = always). With a possible range from 18 to 90, higher scores indicate greater sexual assertiveness. In this sample scores ranged from 33 to 90 (M = 65.8, SD = 12.01). The three subscales included Information Communication (Min. = 6, Max. = 30, M = 24.6, SD = 6.79), Initiation (Min. = 7, Max. = 30, M = 18.4, SD = 4.02), and Refusal (Min. = 6, Max. = 30, M = 21.5, SD = 4.59). The Information Communication subscale investigates a respondents' willingness to request/obtain information about a partner's HIV risk or risk behaviors. The Initiation subscale considers a respondents' propensity to initiate sexual behaviors and the Refusal subscale investigates a respondents' ability to refuse to engage in unwanted sexual behaviors and activities.

Aids-related stigma scale

The AIDS-Related Stigma Scale was developed by Kalichman et al. (2005) and was used as the dependent variable for this analysis. This scale is used to measure stigmatizing beliefs regarding people with AIDS. With nine items using a four-point Likert scale (strongly disagree, disagree, agree, strongly agree), scores can range from 9 to 36. Higher scores indicate higher levels of AIDS-related stigma. Kalichman et al. (2005) reported this scale possesses internal consistency and test–retest reliability. In further analysis, Kalichman et al. (2005) report the instrument possesses convergent and divergent-validity. In this sample, scores ranged from 9 to 26 (M = 13.9, SD = 4.12).

FINDINGS

Description of the Sample (Univariate Analysis)

The sample included 572 women aged 50 to 93, with a mean age of 63.63 years (SD = 10.496). The largest number of respondents identified themselves as married (n = 216, 38.2%). Other respondents indicated that they were widowed (n = 97, 17.2%), divorced (n = 96, 17.0%), living with a partner (n = 67, 11.9%), single (n = 56, 9.9%), have a partner but do not live together (n = 30, 5.3%), or separated (n = 3, 0.5%). In terms of ethnicity, 58.9% (n = 337) self-identified as Non-Hispanic/Non-Latino White, 22% (n = 126) self-identified as Hispanic/Latino, 12.3% (n = 73) self-identified as African American/Caribbean Black, 2.1% (n = 12) self-identified as “other,” and 4.2% (n = 24) did not provide a response. The level of education of the women ranged from 2 to 25 years, with an average education of 14 years. Younger women in the sample had higher levels of education. Most respondents (n = 428, 74.8%) had lived in South Florida for more than 10 years. Approximately one-third of the participants (33.9%) had been tested for HIV and 0.3% (n = 2) of the participants reported a positive result. The majority of the women (88.6%) believed they were not at risk for HIV infection. Participants in this study had received HIV/AIDS information from a variety of sources, including written literature, television, and friends.

Multivariate Analyses

Table 1 reports the bivariate correlations between the dependent variable (HIV-stigma) and other possible predictor variables. To better understand the relationships that might exist between the variable measures with multiple subscales, correlations were computed between the dependent variable and all subscales. As Table 1 evidences, there were significant relationships between the dependent variable and the Rosenberg Self Esteem Scale, the Sensation Seeking Scale, the Self Silencing Scale, and the Sexual Assertiveness Scale. There were significant relationships between all subscales for the Self Silencing Scale (Externalized Self Perception subscale, Care as Self-sacrifice subscale, Silencing the Self subscale, and Divided Self subscale) and the Sexual Assertiveness Scale (Information and Communication subscale, Initiation subscale, and Refusal subscale). No significant correlations existed between age, relationship status, ethnicity, and years in the United States.

TABLE 1 Bivariate Correlations with the Dependent Variable (AIDS Stigma)

Predictor variables	r	p
Rosenberg Self Esteem Scale	−.343	<.001
Sensation Seeking Scale	−.153	<.001
Self Silencing Scale (Total scale score)	.426	<.001
Externalized Self Perception subscale	.347	<.001
Care as Self-sacrifice subscale	.322	<.001
Silencing the Self subscale	.393	<.001
Divided Self subscale	.317	<.001
Sexual Assertiveness Scale (Total scale score)	−.379	<.001
Information and Communication subscale	−.258	<.001
Initiation subscale	−.280	<.001
Refusal subscale	−.251	<.001
Age	.063	.132
Relationship status	.039	.362
Ethnicity	−.006	.888
Years in the United States	−.110	.145

In exploratory analysis all possible predictor variables were entered into a general linear regression. This model possessed an adjusted R square of .181 (F = 4.980, p < .001). However, many of the variables lacked any proximity as significant predictor variables.

In further analysis, variables were excluded from further analysis if they lacked significance. In this continuing analysis an economical model was adopted with the predictor variables of Self Esteem (β = −.152, t = −3.332, p = .001), Sensation Seeking (β = −.079, t = −2.029, p = .001), Self Silencing (β = .240, t = 4.668, p < .001), and Sexual Assertiveness (β = −.193, t = 4.668, p < .001). The model possessed an adjusted r² = .233 (F = 41.432, p < .001). To better understand the components of Self Silencing and Sexual Assertiveness, models were run in which the subscales were used in analysis rather than the total scales. In this analysis, the subscales Externalized Self Perceptions, Divided Self, and Information and Communication were not significant contributors. The final model of this analysis had an adjusted variance of 22.5% (F = 26.854, p < .000). The predictive variables are presented in Table 2 and include: Rosenberg Self Esteem Scale (β = −.190, t = −4.487, p < .000), Care as Self-sacrifice subscale (β = .125, t = 2.714, p = .007), Silencing the Self subscale (β = .164, t = 3.290, p = .001), Initiation subscale (β = .108, t = -2.619, p = .009), and the Refusal subscale (β = −.091, t = −2.154, p = .032). The Sensation Seeking scale was retained in the model in spite of it lacking statistical significant as a cautionary prevention of a Type II error (β = −.073, t = −1.867, p = .062).

TABLE 2 Final Regression Model Summary for AIDS Stigma

Predictor variables	Beta	t	Sig.
Rosenberg Self Esteem Scale	−.190	−4.487	<.001
Sensation Seeking Scale	−.073	−1.867	.062
Self Silencing Scale
Care as Self-sacrifice subscale	.125	2.714	.007
Silencing the Self subscale	.164	3.290	.001
Sexual Assertiveness Scale
Initiation subscale	−.108	−2.619	.009
Refusal subscale	−.091	−2.154	.032

DISCUSSION

This study investigated perceptions of HIV-related stigma among women aged 50 to 93.

Each of the variables in the regression model is considered below. A model was presented with an adjusted variance of 22.5% and included the predictive variables of self esteem, care as self-sacrifice, silencing the self, initiation of sexual activities and behaviors, and refusal to engage in some sexual behaviors and activities. The variable of sensation seeking was retained in the model because of its proximity to significance (p = .062).

HIV-Stigma

Internalized stigma and fear of stigma from potential sexual partners with whom older women address safer sex practices may increase the likelihood of continuing sex behaviors that transmit HIV. In the case of HIV and AIDS, stigma and discrimination related to the way the disease was transmitted (i.e., through sexual contact or drug use) can lead to barriers to testing, fear of disclosure, inadequate and inappropriate treatment, and the further spread of a disease which might otherwise be contained.

To date, there are very few published studies investigating the link between HIV-stigmatizing behaviors and safer sex behaviors of persons who are not infected with the virus. However, previous studies with HIV-infected women for whom stigma management involved efforts to control information to preserve social relations and maintain moral identity (Sandelowski et al., 2004) points to the idea that HIV-stigma may be associated with safer sex behaviors.

Midlife and older women face a double stigma of ageism and the idea that older people should not be having sex, which also acts as a barrier to preventive measures such as condom negotiations and testing (Kalichman et al., 2005) found that individuals (n = 2306) who stated that HIV positive persons should conceal their HIV status had higher AIDS-Related Stigma Scale scores. Also, individuals who refused to report whether they had been tested for HIV scored higher on the AIDS-Related Stigma Scale (Kalichman et al., 2005). Older adults at high risk for HIV infection are less likely to engage in HIV prevention strategies than younger individuals and are less likely to be tested than their younger counterparts. This difference has been thought to be due to the stigma attached to HIV (Emlet & Farkas, 2002; Mack & Ory, 2003).

To further understand midlife and older women in relation to the major variables of this study, the significant correlations between the variables suggested that HIV-stigma scores were the highest in women who possessed lower self esteem, exhibited more self silencing behaviors, and were less sexually assertive.

Self esteem, self silencing, and sexual assertiveness were found to be predictors of midlife and older women's HIV-stigmatizing behaviors. Given the history of women's oppression, older women's economic dependency, fear of abuse, fear of being lonely, and the fact that women place great emphasis on intimate relationships, it could logically be expected that, on the average, midlife and older women would have demonstrated high levels of self silencing and lower levels of sexual assertiveness, even with slightly higher levels of self esteem.

There is some evidence that midlife and older women have difficulty in communicating about sexual behavior (Zablotsky & Kennedy, 2003). To date, there have been no published studies linking self silencing with HIV-stigma or risky sex practices of midlife and older women. Some studies have suggested that the more often women elicit information from their sexual partners, the less likely they engaged in unprotected sex (Quina, Harlow, Morokoff, & Burkholder, 2000). Some studies have shown that women who believed that asking a partner to use condoms implied that he was unfaithful were four times as likely to not use condoms compared with women who did not think that asking a partner to use condoms implied infidelity (Wingood & DiClemente, 2000).

Two subscales (i.e., Initiation of sexual activities and behaviors, Refusal to engage in some sexual behaviors and activities) emerged as predictors. Many women engage in unprotected sex with their male partners with the assumption that their partners do not engage in high risk sexual behaviors (e.g., unprotected sex with men, injection drug use). Older women who hold stigmatizing attitudes toward persons with HIV-infection (which may be due to cultural attitudes about sex and persons from stigmatized groups already associated with HIV, e.g., gay men, drug users, sex workers), may be less comfortable initiating a conversation about their partner's sexual history or high-risk behaviors.

Sensation seeking had no statistical significance as a predictor of HIV-stigma. Sensation seeking, a concept not previously studied in midlife and older women, was significantly correlated with HIV-stigma (p = .000) but was not a predictor in the final model. Sensation seeking has been identified as an important correlate of sexual risk behaviors, and many women are high sensation seekers and an increasing number of women take health risks such as binge drinking (Pirkle & Richter, 2006; Wechsler et al. 2002), which is linked to unsafe sex practices. The nature of the items in the sensation seeking scale chosen for the study questionnaire may not translate well into the context of “sexual” sensation seeking for this sample of midlife and older women. This, combined with the age and gender of the participants and possible absence of moderating variables (i.e., binge drinking), may have influenced the results.

This study has some potential limitations. First, this study used a convenience sample of women aged 50 that was not tested for HIV. Only one-third (33.9%) had been previously tested; two women reported a positive result. The HIV status of the rest of the sample is unknown, making it difficult to ascertain differences related to HIV serostatus that might affect how prevention and intervention strategies are designed and implemented. Second, lack of multiple measures used to measure the study constructs (i.e., Jack and Dill have created the only known reliable instrument that measures self silencing and Kalichman et al.’s AIDS-stigma measure is currently the only reasonable instrument developed to measure HIV-stigma in persons who are not HIV-infected). Third, a cross-sectional, correlational survey design was used and therefore, generalizations cannot be made regarding changes or trends over time, directionality of influence, or cause-and-effect relationships. Fourth, a convenience sample was used to collect data which may produce self-selection effects. Fifth, self-report questionnaires were used for data collection that might have resulted in response bias, social desirability bias, and inaccuracies that could affect findings. Lastly, a brief but significant lapse in data collection occurred due to a series of hurricanes (i.e., Hurricanes Katrina and Wilma, respectively). These events, which dramatically affected the communication, mobility, and emotional well being of South Florida residents, disabled data collection efforts for several weeks.

The results of the final model of this analysis were found to be significant, albeit the overall adjusted variance is relatively small (less than 23%). Further investigation is needed to explore how certain psychosocial factors might mediate or moderate HIV-stigma, and how HIV-stigma impacts the sexual decision-making processes of women aged 50 and older.

IMPLICATIONS FOR SOCIAL WORK PRACTICE AND RESEARCH

Although many advances have been made in the fight to curtail the epidemic, including the advent of highly active antiretroviral treatment (HAART), HIV infection and AIDS continue to be a challenge in terms of developing effective educational strategies and prevention interventions. The increasing impact on midlife and older women calls for a creative and multidimensional approach to preventing the spread of the disease, including policies that give them a voice in HIV/AIDS legislation and program direction (Jacobs & Kane, 2009).

Age and aging present specific challenges to HIV prevention, intervention, and research. Our knowledge about HIV-stigma and how it intersects with age and gender is limited. The biologic, interpersonal, and social properties that make up midlife and older women's life experiences create pathways to HIV acquisition not yet explored. The misconception that women age 50 and older are rarely at risk further complicates the processes of both identifying risk factors and creating effective prevention strategies. Without a vaccine, prevention is paramount in eradicating HIV and AIDS.

It is important to recognize the change in sexual attitudes that has transpired in the last 50 years. American society has shifted from a climate in which explicit reference to sex and sexuality were not spoken aloud to a society in which magazines, radio, movies, and television shows air private issues in great detail. This greater societal tolerance of sexual issues has affected many midlife and older women and will continue to affect the next generation of midlife and older women.

It has been difficult to determine if midlife and older women comply or even are aware of HIV prevention practices due to the stigma attached to HIV and the groups associated with the disease (i.e., gay men, drug users, and sex workers). With the increasing numbers of HIV cases in midlife and older women it is imperative that we clearly identify risk and protective factors that are amenable to intervention.

The awareness that midlife and older women with low levels of self esteem and sexual assertiveness and high levels of self silencing were more likely to report HIV-stigmatizing behaviors is significant. Critical attention should be paid to age and the factors that influence HIV-stigma (i.e., self esteem, self silencing, and sexual assertiveness) when designing HIV prevention strategies.

Considerations of the interactions that take place within intimate partnerships cannot ignore the existence of power relations based on gender, which are characterized by domination, control, and pressures exerted by the male partner. Skill building is needed that focuses on self expression and communication styles used by midlife and older women to negotiate safer sex. The tensions inherent in these relationships must also be addressed in every intervention, emphasizing the nuances of masculine and feminine identities such as sexist stereotypes and gender roles. However, the power relations based on gender inequities that render some midlife and older women vulnerable to HIV infection because of their influence on their social conditions (i.e., economic dependence, poverty, violence, abuse) cannot be avoided.

Gender and generational issues need to be considered in crafting HIV prevention efforts. Older women may not be comfortable disclosing their sexual behaviors to social work practitioners or health care providers due to HIV-stigma. This can make it difficult for both parties to discuss risks and protective strategies. Few Americans over age 50 who are at risk for HIV infection either use condoms or get tested for HIV (Emlet & Farkas, 2002; Mack & Ory, 2003). Physicians are much more likely to rarely or never to ask patients over 50 about HIV risk factors (Skiest & Keiser, 1997).

Many older people live in assisted living communities where there is still great stigma attached to HIV/AIDS, often associated with homosexuality and/or substance abuse. Management may be resistant to providing HIV educational materials or presentations in their facilities. It is important to reduce the stigma surrounding the sexual needs of older women by encouraging them to discuss these issues with their social workers, counselors, health care providers, and their families.

The nation's baby boomers, in general, are a population that is more physically active and healthier than previous generations. Consequently, they may be more sexually active compared to past generations. A low perception of susceptibility to HIV and high levels of stigmatizing attitudes may lead to more sexual risk behaviors. The emphasis on skills building to increase effective and direct communication strategies (i.e., decrease self silencing and increase self esteem and sexual assertiveness) may help decrease HIV-stigmatizing attitudes in older women. Providing HIV education, creating awareness of risk factors, and changing midlife and older women's low perception of risk for HIV can also help decrease the stigma attached to the disease. Designing and implementing successful HIV prevention efforts that focus on behaviors that reduce the risk of HIV infection/reinfection and enhancing protective factors in midlife and older women will require an understanding of the interplay of all these factors.

Research efforts should include interventions that address interpersonal communication and safer sex negotiations, self esteem enhancement, sexual assertiveness training, and HIV-stigma reduction through education to determine their effectiveness in preventing HIV. Ageist attitudes and stereotypes about midlife and older women create barriers in the contexts of social science research. Researchers need to understand how ideas about age and aging shape the types of questions asked and the assumptions upon which conclusions are made. By looking past the stereotypes and acknowledging that older people continue to become infected in later life and more young people will be aging with HIV, there is a tremendous opportunity to expand research efforts to include midlife and older women.

We thank the volunteer research assistants, our community partners, and above all, the women who participated in this study.