Ssrg.med.miami.edu

Sexual Risk Behaviors in Late Middle Age and Older HIVSeropositive Adults Lourdes Illa Æ Andrew Brickman Æ Gilbert Saint-Jean ÆMarisa Echenique Æ Lisa Metsch Æ Carl Eisdorfer ÆVictoria Bustamante-Avellaneda Æ Mario Sanchez-Martinez Ó Springer Science+Business Media, LLC 2008 Little is known about the sexual behaviors of older adults, although the prevalence of HIV/AIDS israpidly increasing in this population. As part of a larger HIV/AIDS has become a significant national problem among multi-site study examining secondary HIV prevention, we older adults (Maes and Louis The estimated number recruited from an HIV primary care clinic 210 sexually of persons living with HIV/AIDS reported in 33 states and active HIV positive individuals aged 45 and over (125 men, US dependent areas increased from 32% in 2001 to 43% in 85 women) who had engaged in vaginal or anal sex within 2005, including those with an AIDS diagnosis whose num- the past six months. Twenty percent of the participants bers increased from 29% to 34 % (Centers for Disease reported inconsistent use of condoms and 33% had multi- Control [CDC], ). Despite the increasing prevalence of ple sexual partners during the previous six months.
HIV and AIDS in older adults, little is known about their Negative mood and perceived HIV stigma were associated sexual behaviors. Studies, however, show that older adults with inconsistent condom use. In addition, multiple sex are sexually active (Dunn and Cutler ). Results from a partners and higher level of education were related to national survey conducted in 2001 by the National Advisory inconsistent condom use during sex with partners of neg- Council on Aging to examine sexual activity among persons ative or unknown serostatus. These findings indicate that over 60 years indicate that more than 92% of the respondents contrary to current beliefs, sexually active older adults, consider sex an important part of life. Notably, responses similar to younger ones, may be engaging in high risk were similar for men and women. Additionally, 75% of those between 65 and 74 considered themselves sexually active(National Bulletin of the National Advisory Council on Aging ). Research across generations have revealed differences between older adults and their younger coun-terparts in terms of sexual knowledge, risk behaviors, andbiological factors, signifying the importance of age appro- priate interventions that target the specific needs of this Division of Child and Adolescent Psychiatry, Department unique cohort (Orel et al. ). Increased age has been of Psychiatry and Behavioral Sciences, University of Miami associated with having incorrect information concerning Miller School of Medicine, 1695 NW 9th Ave, Suite 1404 H,Miami, FL 33131, USA prevention, including the need to protect oneself during high- risk behaviors regardless of age (Henderson et al. ;Zablotsky ). In contrast to the younger age groups, many A. Brickman Á M. Echenique Á C. Eisdorfer Á individuals over 50 do not consider unprotected sex a high- V. Bustamante-Avellaneda Á M. Sanchez-MartinezDepartment of Psychiatry, University of Miami Miller School risk behavior because many are no longer concerned about contraception (Shaw ). This cohort is less likely to beconcerned about HIV and less likely than their younger counterparts to use condoms. Furthermore, older HIV posi- Department of Epidemiology and Public Health, Universityof Miami Miller School of Medicine, Miami, FL, USA tive sexually active adults are less likely than younger adults to change their sexual behaviors based on HIV knowledge HIV positive diagnosis, being age 45 years or older, and reportedly having engaged in vaginal or anal sex within the Older adults also differ from their younger counterparts last twelve months. Individuals with medical, psychiatric in terms of biological factors related to transmission and or cognitive symptoms that limited their ability to partici- disease process (Mack and Bland ). For example, the pate in the study were excluded. Two hundred seventy- progression from HIV to AIDS and the development of seven HIV-positive individuals were recruited into the AIDS opportunistic infections are faster in older persons study between January 2004 and November 2006. All (Adler and Nagle ). Multiple health problems and age- research participants enrolled in Project ROADMAP par- related physiological changes may make elderly persons ticipated in an informed consent process prior to a baseline particularly vulnerable to HIV infection (Gaeta et al. assessment. All procedures were approved by the Institu- For example, post-menopausal women are at greater risk tional Review Board (IRB) of the University of Miami for HIV infection and re-infection during heterosexual Miller School Of Medicine. Participants were compensated contact because of the fragility of the vaginal mucosa for their time for each assessment with a voucher for a local related to decreased levels of estrogen (Shaw The need for prevention interventions specifically tar- geting HIV positive persons is widely recognized since recent prevention efforts with HIV negative individualsfailed to show effectiveness within the HIV positive popu- Measures were selected from the multisite evaluation lation (Mack and Bland ). Specifically, efforts to target assessment tool developed by the University of California, high risk behaviors for prevention of HIV infection and re- San Francisco, Center for AIDS Prevention Studies, infection may be even more critical in the older population Enhancing Prevention with Positives Evaluation Center because of socio-cultural, biological and behavioral vul- nerabilities. Data are sorely needed to adequately inform thedevelopment of age appropriate secondary HIV prevention interventions. In response, Project ROADMAP (Re-edu-cating Older Adults in Maintaining AIDS Prevention), was Demographic variables assessed included age, gender, developed as part of a multi-site clinical demonstration ethnicity, education, income, relationship status, and sexual project examining the efficacy of secondary prevention behavioral interventions. The intervention is guided by theInformation-Motivation-Behavior Skills (IMB) model of AIDS risk-behavior change, a psychosocial model ofbehavior change developed specifically to address the Medical variables assessed included presence or absence of complexity and interpersonal nature of HIV risk reduction as HAART medication, age at time of diagnosis, number of well as other behavioral issues (Fisher et al. ). Since the years since diagnosis and mode of infection.
initial phase of the study was designed to better understandthe sexual risk behaviors of older HIV positive adults, the data presented in this paper is considered exploratory.
Sexual risk was measured by asking participants informa-tion regarding their sexual behaviors within the previous six months: number of partners, partner’s gender, partner’sHIV status, type(s) of sexual act and condom use. Partic- ipants were asked to report how many oral, vaginal andanal sexual acts they had with each HIV positive, negative Participants were recruited from a primary care clinic in a or unknown serostatus partner and of those acts how many large urban medical center. The University of Miami/ included condom use. For example, ‘‘How many times did Jackson Memorial Medical Center (UM/JMH) is Miami- you have vaginal sex with your HIV-positive partner? Of Dade County’s largest health care facility for HIV/AIDS the X times, how many times was a condom used?’’ treatment, serving 3,500 patients annually. The clinic To these measures we added a local evaluation com- population is ethnically and racially diverse with 57% of ponent. The following variables were selected because they participants Hispanic, 21% White non-Hispanic, and 23% have been associated with HIV transmission/risk behavior Black non-Hispanic. This latter group is comprised of in previous research studies (Kalichman and Nachimson African Americans as well as other Blacks of Caribbean ; Kelly et al. ; Maes and Louis ; Preston descent. Inclusion criteria for the study were having an ). The instrument has 40 items measured on a 4-pointLikert-type response set. The assessment consists of four HIV knowledge was measured using 33 items derived from subscales: personalized stigma, disclosure, negative self CDC’s Handbook for Evaluating HIV Education (CDC, image and public attitudes. The HIV Stigma Scale has been 2000), which we adapted to the 45 year and older popu- found to be reliable and valid with a large, diverse sample lation. A summary score was calculated by summing item of people living with HIV. Reliability was 0.96 for the total scores, with higher scores representing greater knowledge.
scale, 0.93 for the personalized stigma sub-scale, 0.93 forthe disclosure sub-scale, 0.88 for the negative self-image subscale, and 0.93 for the public attitudes subscale. Asummary score was calculated by summing item scores, Sexual self-efficacy was measured with a 7-item self-report with higher scores representing a lower perceived HIV questionnaire measuring respondents’ beliefs about their capacities to engage in safe sex (Malow et al. Per- Measures were administered by a clinical interviewer ceived self-efficacy was measured in terms of judgments of using Computer Assisted Personal Interview (CAPI) soft- personal capabilities and the strength of that belief. A ware. The duration of the assessment session was summary score was calculated by summing item scores, with higher scores representing greater sexual self-efficacy.
Analyses were conducted only for the sub-sample of Mood was measured using the Profile of Mood States individuals who reported having vaginal or anal intercourse (POMS), a measure of transient mood states consisting of within the previous six months (n = 210), referred to 65 items (McNair et al. ). The POMS has six indi- hereby as sexually active. Descriptive statistical analyses vidual mood subscales and one overall index of distress were used to examine the distribution of the sample’s called Total Mood Disturbance (TMD). The POMS sub- demographic, psychosocial, and behavioral variables. We scales have been shown to have internal consistencies near employed univariate logistic regression analyses to evalu- .90 or above and test-retest reliabilities ranging from .65 to ate the unadjusted association of condom use with the other .74 and to have good predictive and construct validity variables and multiple logistic regression to independently (McNair et al., 1971). The POMS has also been used in evaluate these variables in relation to two behaviors: con- previous studies with HIV infected samples and has been sistent use of condoms during sex with all partners and shown to be sensitive to stressor-related affective responses consistent use of condoms during sex specifically with (Antoni et al. Lutgendorf et al. We calculated negative/unknown serostatus partners. The independent summary scores for the overall scale as well as its sub- variables for the logistic regression analysis were: age scales using the guidelines established in the POMS (continuous), gender (male/female), education (Higher Manual, with greater scores representing worse mood for Education/High School or lower), marital status (commit- the total mood scale and all subscales except the vigor ted relationship: married/domestic partnership/committed subscale, where greater scores signified increased vigor.
Available sets of norms for interpretation of scores include adults, geriatric adults, college students and psychiatric partners (yes/no); and the full scales measuring mood, self- outpatients. Although the POMS is not a diagnostic tool, T efficacy, knowledge, and perceived HIV stigma, which scores of 65 or greater are suggested as a common cut- were entered as continuous variables in the models. The point for cases warranting special attention. To determine dependent variable, ‘inconsistent use of condom,’ refers to what percentage of our participants scored above the participants who did not use condoms at least once in the clinical cutoff, we calculated the percentage of participants preceding six months during anal or vaginal sex; these between ages 45 and 54 who scored a T score of 65 or participants were coded 1. Participants who reported using greater, utilizing the adult norms, and the percentage of condoms 100% of the time (consistent use of condom) participants aged 55 and over utilizing the geriatric norms.
were coded 0. We entered into the logistic regressionmodels only variables that were associated with the dependent variables at P \ 0.20 in bivariate analysis(Hosmer and Lemeshow Variables were removed Perceived HIV stigma was assessed using the Perceived from the model through a stepwise backward iterative HIV Stigma Scale, a self-report instrument measuring HIV process if their contribution to the overall model was stigma perceived by people living with HIV (Berger et al.
greater than or equal to a p value of 0.10 (Field (n = 6) for confusion, 44% (n = 15) for decreased vigorand 14.7% (n = 5) for total mood problems.
Of the 210 sexually active participants, 59.5% (n = 125)were men and 40.5% (n = 85) were women. They had a mean age of 51 years (range 45–71, SD = 5.23). Eighty-two percent (n = 172) self-identified as Black, 12% Eighty-four percent (n = 105) of men identified themselves (n = 26) Hispanic, and 5% (n = 11) White. Sixty-two as heterosexual, 8% (n = 10) as homosexual, and 8% percent (n = 131) reported a High school education, of (n = 10) as bisexual. Forty-six percent (n = 48) reported which 24% (n = 50) had some higher education. Of our having more than one partner; 5% had 10 or more partners.
participants, 89% (n = 187) reported a yearly income of Sixty percent (n = 75) of this HIV-positive sample had sex $10,000 or less. Only 15% (n = 31) of our sample reported at least once with a negative or unknown status partner.
being in a committed relationship (married, domestic Eighty-seven percent (n = 109) had engaged in vaginal sex partnership or committed relationship not living together).
and 20% (n = 25) in anal sex. The median number of The majority of participants, 94% (n = 197), reported sexual acts within the last six months was 6 and the range being infected through heterosexual contact. The mean age was 1–240. When asked to report the gender of their sexual at the time of HIV diagnosis was 39.7 years (range 23–69, partners during the previous six months, 85.6% (n = 107) SD = 8.58), with 25% (n = 52) diagnosed at 45 or older.
of the men indicated they had sex exclusively with women, The mean time since diagnosis was 11 years (range 0–22, 12% (n = 15) exclusively with men, and 2.4% (n = 3) SD = 6.17). Ninety-two percent (n = 193) of participants reported having sex with both men and women. No self- were on antiretroviral therapy (ARV) at the time of identified heterosexual men reported having sex with men POMS scores are reported in Table . Among participants Of the 85 women, 95% (n = 80) identified themselves as aged 45 to 54, 9.3% (n = 14) scored in the at risk range for heterosexual, 1% (n = 1) as homosexual, and 5% (n = 4) depression, 10% (n = 15) for tension/anxiety, 8.6% as bisexual. Fourteen percent (n = 12) reported having (n = 13) for anger, 8% (n = 12) for fatigue, 12% (n = 18) more than one partner and 2% (n = 2) reported more than for confusion, 27% (n = 41) for decreased vigor and 11.3% 14 partners. All of the women engaged in vaginal sex and (n = 17) for total mood problems. Among those aged 55 5% (n = 4) in anal sex. Their median number of sexual and older, 20.5% (n = 7) scored in the at risk range for acts in the previous six months was 7 (range: 1–202). When depression, 14.7% (n = 7) for tension/anxiety, 26.5% asked to report on their sexual behavior with men within (n = 9) for anger, 11.8% (n = 4) for fatigue, 17.6% the last six months, 61% (n = 52) mentioned they had sex Table 1 POMS mean scores and percent with scores above norms, adult norms and geriatric norms Note: Test statistic: One Sample t-test. P values represent differences between sample means and normative mean values at least once with a partner whose HIV status was negative Specifically, of the POMS subscales, depression (OR = 1.05, P \ 0.01), tension/anxiety (OR = 1.06, P \ 0.10), anger/hostility (OR = 1.06, P \ 0.01), fatigue (OR = 1.06, P \ 0.05) and confusion (OR = 1.10, P \ 0.01) correlatedsignificantly with condom use. On multivariate analysis (Table only negative mood emerged as a predictor ofinconsistent condom use (OR = 1.01, P \ 0.01).
Almost 20% (n = 41) of the sexually active participantsreported not using condoms consistently. On bivariate anal- Sex with Negative/Unknown Serostatus Partners ysis (Table ), no medical or demographic variables,including gender and sexual orientation, and neither self- Sixty percent (n = 127) of participants who reported hav- efficacy nor knowledge was found to be associated with ing vaginal or anal sex engaged in sexual activity at least condom use. In contrast, greater perceived HIV stigma (OR = once with negative/unknown serostatus partners. Of these, 0.98, P \ 0.05) and overall negative mood (OR = 1.01, 17.3% (n = 22) reported not using condoms consistently.
P \ 0.01) were associated with inconsistent condom use.
No significant relationship was found between unprotectedsex with negative or unknown serostatus partners and age, Table 2 Association of inconsistent condom use during sexual gender, age at diagnosis, time since diagnosis, HIV knowledge, and sexual self-efficacy. Inconsistent condomuse during sex with negative/unknown serostatus partners also had a significant positive association with multiple sexual partners (OR = 2.20, P \ 0.10), post High schooleducation (OR = 2.81, P \ 0.05), perceived HIV stigma (OR = 0.97, P \ 0.05) and negative mood (OR = 1.01, P \ 0.05) on bivariate analysis (Table ). Of the POMS subscales, depression (OR = 1.04, P \ 0.05), and anger/ hostility (OR = 1.05, P \ 0.10) correlated significantly with condom use. On multivariate analysis (Table inconsistent condom use was positively associated with negative mood (OR = 1.02, P \ 0.05), multiple partners (OR = 2.69, P \ 0.05) and post High school education (OR = 3.34, P \ 0.05). Further exploratory analyses revealed 21.2% shared variance between negative mood and perceived HIV stigma (r = 0.46, P \ 0.001) which may partly explain why perceived stigma was not signifi- cantly related to condom use after controlling for mood.
This study explores sexual behavior among an older sam- * P \ 0.20, ** P \ 0.10, *** P \ 0.05, **** P \ 0.01 ple of sexually active HIV positive adults. While our Table 3 Logistic models for inconsistent condom use during intercourse Note: Variables entered into logistic model for sex with all partners were negative mood and perceived HIV stigma. Variables entered intologistic model for sex with negatives/unknown partners were age, higher education, negative mood, multiple partners and perceived HIV stigma findings are not reflective of the general older HIV positive consistent with findings in the literature indicating a pos- population since participants had to be sexually active sible relationship between stigma and negative mood (Miles within the previous year, they do indicate that among those et al. ). These results suggest that therapeutic inter- who are sexually active, many engage in high risk trans- ventions targeting mood state may improve condom use.
mission behaviors. Twenty percent of the sexually active Such interventions may draw from empirically based HIV positive older adults reported inconsistent condom use treatments for depression in HIV positive persons, such as during the previous six months. Additionally, the findings cognitive behavioral stress management (Cruess et al. ; that only 15% of our HIV positive older adults were in a Lechner et al. and effective secondary prevention committed relationship and that 33% of the sample had interventions that address condom use skills including multiple sexual partners during the previous six month effective condom negotiation (Kalichman et al. ).
period are notable. This indicates that, contrary to norma- Finally, individuals who were educated beyond high tive beliefs, the majority of older sexually active HIV school were more likely to engage in inconsistent condom positive individuals may be having sex in uncommitted use. This finding is counter-intuitive and should be further relationships. Similar to data from younger HIV positive explored given that we have very limited data on socio- adults, having multiple partners was independently related to inconsistent condom use among sexually active HIV Our sample may not be representative of the older HIV positive older adults during sex with partners of negative or positive national population, therefore possibly limiting the unknown serostatus (Eich-Hochli et al. ; Heckman generalizability of our findings. All participants in our et al. Older adults are often confronted by ageism, sample were recruited from a primary care clinic serving even among their health care providers. Studies have primarily individuals of low socio economic status and of demonstrated that the majority of physicians do not discuss African American, Hispanic or White background. Our sexual behaviors with their established HIV positive sample was predominantly heterosexual and not active patients of any age (Metsch et al. ; Morin et al. drug users. Findings do not speak to the risk behaviors of Providers may be particularly reluctant to directly discuss HIV positive older adults who are not seeking medical sexual matters and HIV risk behaviors with older adults care. The results were also based on self-report data, which (Gott et al. ; Lindau et al. ). The CDC (1997) depend on the veracity of the respondents. Although there reports that of the 94% of people age 50 and older who is evidence suggesting that 3, 6 and 12 month recall is less have seen their doctors in the previous five years, only 15% biased than shorter periods and that longer time frames are had discussed HIV risks with their physicians and 72% of more likely to be representative of an individual’s behavior these discussions occurred at the patient’s request. Thus, (Schroder et al. the reader is cautioned since the health care providers should be aware that sexually active validity and reliability of retrospective recall is often a HIV positive older adults may be engaging in risky trans- concern. However, such limitations are inherent in psy- chosocial studies and researchers continue to rely on self- Previous literature suggests that older HIV positive report methods to assess sexual behavior since ethical and adults are more vulnerable to disturbances of mood (Rabkin practical considerations limit the use of more direct et al. ). Similarly, in our study, participants aged 55 and assessment methods (Weinhardt et al. ). Recall may over were more likely to score above the at risk-cutoff point have been particularly an issue for this older sample given for mood symptoms than those between the ages of 45 and the expected challenge to recollect and describe intimate 54. We found an association between negative mood and behaviors that had happened during the preceding six inconsistent condom use in our sample. Further analysis month period. The cross-sectional nature of the data pre- revealed that within the POMS, the depression, tension/ vented the determination of any temporal relationship anxiety, anger/hostility, fatigue and confusion subscales between the predictor variables and the outcomes, thereby correlated significantly with inconsistent condom use precluding a causal interpretation of the observed associ- independent of partner serostatus. The depression and ations. Finally, we are reporting baseline data for an HIV anger/hostility subscales were associated with inconsistent prevention study and it is possible that individuals who condom use with partners of negative or unknown sero- were more concerned about condom use self-selected into status. These findings are consistent with previous research the study. Nevertheless, the study findings represent a describing the association between unprotected sex and contribution to the important issue of understanding the negative affective states (Marks et al. ) and loneliness sexual behaviors of sexually active HIV positive older and instability in self-esteem scores (Martin and Knox adults. The data suggests that sexually active HIV positive ). Perceived HIV stigma was not associated with older adults are engaging in high risk HIV transmission inconsistent condom use in multivariate analysis, probably behaviors and that negative mood states may be important because of its shared variance with negative mood. This is predictors of these behaviors in this population. It emphasizes, too, the need for mental health screening in the Kalichman, S. C., Rompa, D., Cage, M., DiFonzo, K., Simpson, D., primary care setting. This information may be used in Austin, J., Luke, W., Buckles, J., Kyomugisha, F., Benotsch, E.,Pinkerton, S., & Graham, J. (2001). Effectiveness of an health care settings to better address the needs of this intervention to reduce HIV transmission risks in HIV-positive unique population and to develop interventions to help people. American Journal of Preventive Medicine, 21, 84–92.
older HIV positive adults engage in safer sexual practices Kelly, J. A., Murphy, D. A., Bahr, R., Kalichman, S. C., Morgan, M.
both to maintain their own health and to prevent the spread G., Stevenson, Y., Koob, J. J., Brasfield, T. L., & Bernstein, B.
M. (1993). Outcome of Cognitive-behavioral and support group brief therapies for depressed, HIV-infected persons. AmericanJournal of Psychiatry, 150, 1679–1686.
Special Projects of National Significance (SPNS) Lechner, S. C., Antoni, M. H., Lydston, D., LaPerriere, A., Ishii, M., Initiative Funded by the Health Research and Services Administration Devieux, J., Stanley H., Ironson G., Schneiderman, N., Brondo- lo, E., Tobin J. N., & Weiss S. (2003). Cognitive-behavioralinterventions improve quality of life in women with AIDS.
Journal of Psychosomatic Research, 54(3), 253–261.
Lindau, S. T., Leitsch, S., Lundberg, K., & Jerome, J. (2006). Older women’s attitudes, behavior, and communication about sex andHIV: A community-based study. Journal of Women’s Health, 15, Adler, W. H., & Nagel, J. E. (1994). Acquired immunodeficiency syndrome in the elderly. Drugs and Aging, 4, 410–416.
Lutgendorf, S. K., Antoni, M., Ironson, G., Costello, N., Zuckerman, Antoni, M., August, S., Baggett, H. L., Ironson, G., LaPerriere, A., M., Klimas, N., Fletcher, M., & Schneiderman, N. (1998).
August, S., Klimas, N., Schneiderman, N., & Fletcher, M. A.
Changes in cognitive coping skills and social support during (1991). Cognitive-behavioral stress management intervention cognitive behavioral stress management intervention and distress buffers distress responses and immunologic changes following outcomes in symptomatic human immunodeficiency virus notification of HIV-1 seropositivity. Journal of Consulting and (HIV)––seropositive Gay Men. Psychosomatic Medicine, 60, Berger, B. E., Ferrans, C., & Lashley, F. (2001). Measuring perceived Mack, K. A., & Bland, S. D. (1999). HIV testing behaviors and HIV stigma in people with HIV: Psychometric assessment of the attitudes regarding HIV/AIDS of adults aged 50–64. Gerontol- HIV Perceived HIV stigma Scale. Research in Nursing and Maes, C., & Louis, M. (2003). Knowledge of AIDS, perceived risk Centers for Disease Control and Prevention. (2005). Cases of HIV of AIDS, and at-risk sexual behaviors among older adults.
Infection and AIDS in the United States and Dependent Areas, Journal of The American Academy of Nurse Practitioners, 15, 2005. HIV/AIDS Surveillance Report, 17.
Cruess, S., Antoni, M., Hayes, A., Penedo, F., Ironson, G., Fletcher, Malow, R., Devieux, J., Jennings, T., Lucenko, B., & Kalichman, S.
M., Lutgendorf, S., & Schneiderman, N. (2002). Changes in (2001). Substance-abusing adolescents at varying levels of HIV mood and depressive symptoms and related change processes risk: Psychosocial characteristics, drug use, and sexual behavior.
during cognitive-behavioral stress management in HIV-infected Journal of Substance Abuse, 13, 103–117.
men. Cognitive Therapy and Research, 26, 373–392.
Marks, G., Scott, B., & Peterman, T. (1999). Reducing sexual Dunn, M., & Cutler, N. (2000). Sexual issues in older adults. AIDS transmission of HIV from those who know they are infected: The need for personal and collective responsibility. AIDS, 13(3), Eich-Hochli, D., Niklowitz, M., Clement, U., Luthy, R., & Opravil, M. (1998). Predictors of unprotected sexual contacts in Martin, J. I., & Knox, J. (1997). Self-esteem instability and its HIV-infected persons in Switzerland. Archives of Sexual implications for HIV prevention among gay men. Health and Field, A. (2005). Discovering statistics using SPSS (2nd ed.). London: McNair, D. M., Lorr, M., & Droppleman, L. F. (1981). Profile of mood states (POMS) manual. San Diego, CA: Educational and Fisher, J., Misovich, S., Kimble, D., & Weisntein, B. (1999).
Dynamics of HIV Risk Behavior in HIV-Infected Injection Drug Metsch, L., Pereyra, M., Rio del C., Gardner, L., Duffus, W., Users. AIDS and Behavior, 3, 41–57.
Dickinson, G., Kerndt, P., Mahoney Anderson, Strathdee, S., & Gaeta, T., LaPolla, C., & Melendez, E. (1996). AIDS in the elderly.
Greenber, A. (2004). Delivery of HIV Prevention counseling by Journal of Emergency Medicine, 14, 19–23.
physicians at HIV medical care settings in 4 US cities. American Gott, M., Hinchliff, S., & Galena, E. (2004). General practitioner Journal of Public Health, 94, 1186–1192.
attitudes to discussing sexual health issues with older people.
Miles, M. S., Burchinal, P., Holditch-Davis, D., Wasilewski, Y., & Social Science and Medicine, 58, 2093–2103.
Christian, B. (1997). Personal, family, and health-related corre- Heckman, T., Kelly, J., & Somlai, A. (1998). Predictors of continued lates of depressive symptoms in mothers with HIV. Journal of high-risk sexual behavior in community sample of persons living with HIV/AIDS. AIDS and Behavior, 2, 127–135.
Morin, S., Koester, K., & Maiorana, A. (2004). Missed opportunities: Henderson, S., Bernstein, L., St. George, D., Doyle, J., Anuradha S., Prevention with HIV-infected patients in clinical care settings.
& Paranjape, G. (2004). Older women and HIV: How much do Journal of Acquired Immune Deficiency Syndrome, 36(4), 960– they know and where are they Getting Their Information? Journal of the American Geriatrics Society, 52, 1549–1553.
National Bulletin of the National Advisory Council on Aging. (2002).
Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression Sex over sixty. Expression: Bulletin of The National Advisory (2nd ed.). New York: Wiley- Interscience.
Kalichman, S. C., & Nachimson, D. (1999). Self-Efficacy and Orel, N., Spence, M., & Steele, J. (2005). Getting the message out to Disclosure of HIV-positive Serostatus to Sex Partners. Health older adults: Effective HIV education risk reduction publica- tions. Journal of Applied Gerontology, 24, 490–500.
Preston, D. B., D’Augelli, A. R., Kassab, C. D., Cain, R. E., Schulze, Shaw, J. K. (2001). HIV and the older adult: A hidden epidemic. HIV F. W., & Starks, M. T. (2004). The influence of stigma on the sexual risk behavioral of rural men who have sex with men.
Weinhardt, L. S., Forsyth, A. D., Carey, M. P., Jaworski, B. C., & AIDS Education and Prevention, 16, 291–303.
Durant, L. (1998). Reliability and Validity of self-report Rabkin. J., McElhiney, M., & Ferrando, S. (2004). Mood and substance measures of HIV-related sexual behavior: Progress since 1990 use disorders in older adults with HIV/AIDS: Methodological and recommendations for research and practice. Archives of issues and preliminary evidence. AIDS, 18(suppl 1), S43–S48.
Schroder, K., Carey, M., & Vanable, P. (2003). Methodological Zablotsky, D. L. (1998). Overlooked, ignored, and forgotten: Older challenges in research on sexual risk behavior: II. Accuracy of women at risk for HIV infection and AIDS. Research on Aging, self reports. Annals of Behavioral Medicine, 26, 104–123.

Source: http://www.ssrg.med.miami.edu/documents/Illa-%20Sexual%20Risk%20Behaviors.pdf