Background: There is evidence that a high proportion of people living with HIV infection are unaware of their serologic status, which indicates unmet prevention needs. The aim of the present study was to estimate the rate of lifetime HIV testing and assess the association between previous HIV testing and sociodemographic, clinical characteristics, and risk behaviors in psychiatric patients.
Methods: Cross-sectional national multicenter study including 2380 chronic mentally ill patients randomly selected from 26 Brazilian mental health institutions. Multivariate logistic regression analysis was used to describe the association between patient characteristics and previous HIV testing.
Results: The rate of previous lifetime HIV testing was 27%. Most participants were female (52%), 40 years old or younger (52%) with low schooling (59%), and had previous psychiatric hospitalizations (57%) and schizophrenia as a primary condition (48%). Being tested for HIV was independently associated (P < 0.05) in the multivariate analysis with higher schooling, being younger than 40 years old, living alone, having other self-reported medical conditions, higher HIV/AIDS knowledge, substance use as a main psychiatric condition, a history of sexually transmitted diseases, sexual abuse, and inconsistent lifetime condom use. However, those who never had sexual intercourse and did not know their HIV risk were less likely to be tested.
Conclusions: The rate of HIV testing among these psychiatric patients was low despite their increased risk for HIV infection. These findings support the need for increasing HIV testing and awareness of HIV status among psychiatric patients. Mental health providers should play an active role in providing HIV testing and counseling to this population.
From the *Research Group in Epidemiology and Health Services Evaluation (GPEAS/CNPq), Department of Preventive and Social Medicine, Medical School, Universidade Federal de Minas Gerais, Belo Horizonte (MG), Brazil; †Raul Soares Institute and Teaching Hospital, State Health Department (FHEMIG), Belo Horizonte (MG), Brazil; ‡Department of Demography, Center for Development and Regional Planning, Universidade Federal de Minas Gerais, Belo Horizonte (MG), Brazil; and §Department of Preventive and Social Medicine, Medical School, Universidade Federal de Minas Gerais, Belo Horizonte (MG), Brazil.
This work was carried out by the Federal University of Minas Gerais with technical and financial support of the Ministry of Health/Secretariat of Health Surveillance/Department of STD, AIDS and Viral Hepatitis through the Project of International Technical Cooperation 914/BRA/1101 between the Brazilian Government and the United Nations Educational, Scientific and Cultural Organization-UNESCO.
The authors have no conflicts of interest to disclose.
Correspondance to: Ana Paula Souto Melo, ScD, MD, Department of Preventive and Social Medicine, School of Medicine, Universidade Federal de Minas Gerais, Av. Prof. Alfredo Balena 190, Bairro Santa Efigenia, CEP 30130-100 Belo Horizonte, MG, Brazil. (e-mail: email@example.com).
Evidence shows that a high proportion of people living with HIV infection are unaware of their status, approximately one in five (United States) and one in three (United Kingdom).1,2 This finding points to substantial unmet prevention needs and can be used to focus on HIV testing and other prevention initiatives, particularly in vulnerable populations. Difficult access to HIV testing is a major reason for delayed antiretroviral therapy initiation. Recent studies in Brazil have shown that a high proportion of patients initiate antiretroviral therapy at an advanced stage of disease, indicating the need to develop strategies that can increase the rate of early HIV diagnosis nationwide.3
The prevalence of HIV infection is higher among psychiatric patients than in the general population with rates in developed and in developing countries ranging from 0 to 29.0.4-6 In addition, the rates of sexual and substance use behaviors in patients with severe mental illness associated with HIV transmission, including unprotected intercourse, multiple partners, sex trading, and injection drug use, are of public health concern.6,7 Specific efforts are needed to improve knowledge of HIV status among psychiatric patients focusing on HIV testing as a key component of HIV prevention and early treatment.
Studies about HIV testing in psychiatric patients have mostly focused on more severe patients among selected populations (eg, hospitalized) and are usually based on small sample sizes.8,9 In the United States, rates of lifetime HIV testing ranged from 11.0% to 89.0%.8,9 Consistent correlates of HIV testing include HIV risk behaviors, substance abuse, and higher level of social support. The same correlates have been described in psychiatric population but the results are inconsistent.6,8 HIV testing in this population is not proportional to the elevated levels of infection and potential risk of transmission. In Brazil, HIV testing is voluntary, confidential for all individuals, and it is available throughout the public health services. The only published study that assessed lifetime HIV testing among those with severe mental illness in Brazil indicated a prevalence of 54.1%10 and there are no published studies on the correlates with HIV testing in this population. We should note that AIDS and psychiatric care services in Brazil are public and universal, and yet there is little integration among them.11 Thus, the aim of the present study was to estimate the rate of lifetime HIV testing and assess the association between previous HIV testing and sociodemographic and clinical characteristics and risk behaviors in a representative sample of psychiatric patients in Brazil.
A cross-sectional multicenter study (PESSOAS Project) was conducted in 11 public psychiatric hospitals and 15 public mental health outpatient clinics (CAPS) in Brazil in 2006 to 2007. PESSOAS Project's main objective was to estimate HIV, syphilis, and hepatitis B and C seroprevalence and assess risk behaviors in psychiatric patients as detailed elsewhere.5,7,11
A two-stage probability sampling was used proportional to the type of care (hospital or outpatient care center [CAPS]) and distribution of reported AIDS cases by Brazilian region. First, mental health services were randomly selected proportionally to the number of beds in the hospital stratum or number of registered patients in the CAPS stratum in each region. The second sampling stage was carried out by selecting patients using simple random probability process at each site.11 Eligibility criteria included adult (18 years old or older) psychiatric patients receiving care either at hospitals or adult CAPS. Only patients who were considered capable of providing written informed consent and understood the aims of the study were included in the study. This assessment was based on a preliminary evaluation adapted from the Mini-Mental Status Examination by trained mental health professionals (brief-MMSE). In addition, the presence of acute psychosis was ascertained using a qualitative assessment. Public mental health outpatient clinics that exclusively treated substance use disorders as primary diagnoses were also excluded from the sampling process as a result of the possibility of overestimating selected risk behaviors and/or prevalence rates. The study was approved by the Research Ethics Committee of Universidade Federal de Minas Gerais (UFMG/ETIC 125/03) and the Brazilian Review Board (CONEP 592/2006).
Exposure and Event Measurements
A semistructured person-to-person interview was conducted to collect sociodemographic, clinical and behavioral data, including information on HIV testing. All interviews were carried out by experienced mental health providers. The study protocol, questionnaires, and procedures were tested in a pilot study, as previously described.12 The outcome measure of interest in this study was prior HIV testing ever. Potential explanatory characteristics investigated included: 1) sociodemographic information (gender, age, marital status, racial/ethnic category, literacy, schooling, income in the past 6 months, health insurance, homelessness, history of incarceration, and religion); 2) psychiatric conditions (previous psychiatric admissions, institutionalization, presence of delusion/delirium during the interview, brief MMSE score, main psychiatric diagnoses obtained from medical charts and grouped according to the International Classification of Diseases; and 3) risk behaviors (use of alcohol in the last month, use of crack in the last year, lifetime condom use, HIV/AIDS knowledge, history of sexually transmitted diseases [STDs], self-perception of HIV risk, and lifetime sexual violence). Lifetime condom use was categorized as always, inconsistent use (most of the times and rarely), never, and never had sexual intercourse. HIV knowledge was assessed based on a mean score of 10 questions rated on a 0 to 10 scale.13 The brief-MMSE score was based on seven questions from the MMSE (range, 0-7) on orientation (four questions), memory (two questions), and attention (one question). Self-perception of HIV risk was how the participants assessed their risk of becoming HIV-infected (none, low-, medium-, high-risk, and do not know). Reliability of the interview was considered adequate as previously published.5,12
A descriptive analysis was carried out and Pearson chi-square test was performed for the analysis of categorical data. Lifetime HIV testing was estimated by dividing the number of participants that reported lifetime HIV testing by the total number of participants. These estimates and their related 95% confidence intervals were adjusted for within-cluster correlation considering each mental health service as a potential cluster. Point estimates were also proportionally weighted by sample size of each site relative to its total population, ie, number of beds or number of registered patients. The magnitude of the associations between putative risk factors and history of HIV testing was given by odds ratios and 95% confidence intervals. The level of significance was 0.05. A logistic regression was conducted to identify independent correlates of HIV testing. Variables that were significant in the univariate analysis (P < 0.20) were included in the multivariate model. Variables were entered in a series of blocks in the following order: demographics, clinical psychiatric factors and type of mental health treatment settings, and risk behavior. Forward deletion was used and the Wald test was used to assess the statistical importance of each variable. Only variables with P values < 0.05 remained in the final logistic regression model. Goodness of fit of the final model was assessed by Hosmer-Lemeshow test. STATA 10 (STATA Corp, College Station, TX) was used for data analysis and Paradox Windows was used for database management (Corel Corporation, Ottawa, Canada).
Of 3255 patients recruited, 2763 (84.9%) were eligible to participate in the study. Of these, 2475 were interviewed (89.6%) and 288 were nonparticipants (10.4%). The main reasons for nonparticipation were refusals (52.0%), not being located (19.1%), missed appointments (8.0%), not eligible (3.8%), legally incapable (1.0%), death (0.7%), and miscellaneous reasons (15.4%). Finally, among those interviewed, 2380 (96.2%) had available information on HIV testing for the current analysis. No statistically significant differences were seen between participants and nonparticipants regarding age, gender, schooling, or psychiatric diagnosis (P > 0.05).5
Of the 2380 participants with information available on history of HIV testing, 668 (26.9%; 95% confidence interval, 19.0-34.0%) had been previously tested. They were mostly recruited from CAPS (64.8%), female (51.8%), younger than 40 years of age (51.7%), single (48.0%), and their average individual monthly income was $210 US (below the monthly minimum wage in Brazil at the time of the study). More than half of the sample had less than 5 years of schooling (58.9%) and 17.1% were illiterate. Approximately 7.2% were living in the hospital, whereas 14.3% reported living alone and 17.9% had a history of homelessness. Schizophrenia and other psychotic disorders were the most common conditions (47.7%) followed by depression (12.8%), bipolar disorder (9.0%), substance abuse (7.0%), and anxiety (3.6%). Fifty-eight percent had at least one previous psychiatric hospitalization. One third of the participants (35.0%) had never used condoms, 45.9% reported inconsistent condom use (most of the time or rarely), 7.6% always used condoms, and 11.5% never had sexual intercourse. In addition, 23.2% reported previous STDs, and 59% perceived themselves as at no risk for HIV infection. Lifetime verbal, physical, or sexual violence was reported by 68.9%, 58.0%, and 19.9%, respectively, and 25.4% reported prior incarceration.
Univariate and Multivariate Analyses
Table 1 shows the sociodemographic variables studied. The proportion of participants who had ever been tested for HIV was significantly higher among those younger than 40 years old with higher schooling and private health insurance, who were living alone, and had a history of homelessness or incarceration. Similarly, among the clinical related variables, a higher rate of lifetime HIV testing was also seen among those who reported other medical conditions, substance abuse as their main psychiatric condition, and higher brief MMSE scores. Those institutionalized in psychiatric hospitals had statistically lower rates of HIV testing. However, there were no differences in rates of HIV testing between treatment settings (hospitals or CAPs). Finally, regarding risk behaviors, higher rates of HIV testing were found among those reporting use of alcohol in the last month, use of crack in the previous year, higher HIV/AIDS knowledge, history of STDs, and a history of sexual abuse. Lower rates of lifetime HIV testing was found among those who perceived themselves at no risk or did not know their HIV risk, those reporting never using condoms during their lifetime, or never having had sexual intercourse.
Final multivariate analysis showed 10 variables that were independently associated with lifetime HIV testing (P < 0.05). Higher schooling, being younger than 40 years old, living alone, having other self-reported medical conditions, higher HIV/AIDS knowledge, substance use as the main psychiatric condition, history of STDs, sexual abuse, and inconsistent lifetime condom use were all associated with higher testing rates, whereas never having had sexual intercourse and not knowing their HIV risk were associated with lower HIV testing ever (Table 2).
The results of the present study show that the rate of previous HIV testing (26.9%) in psychiatric patients was lower than the general adult Brazilian population in 2005 (33.6%).14 This prevalence of HIV testing found in the present study is also very low compared with other studies in psychiatric patients in the international literature and the single Brazilian study.6,8,10,15,16
HIV testing has been increasing worldwide through general population campaigns. One of the core components of Centers for Disease Control and Prevention HIV prevention strategy is to increase the number of HIV-infected persons who know their serostatus.17 In Brazil, the Ministry of Health has also launched a similar campaign with the same goal.18 Recently, França-Júnior et al14 reported an increase in HIV testing in Brazil from 20.0%, in 1998 to 33.6% in 2005. However, the lower rate of HIV testing found in Brazilian psychiatric population is of public health concern because psychiatric patients have well-documented higher rates of HIV prevalence and risk behaviors for STDs as compared with the general population.4,6
In general, our results are consistent with those in the international literature, especially regarding risk behaviors. We found that participants who reported history of STDs and sexual abuse and inconsistent condom use had higher rates of HIV testing. Most population-based studies and other studies carried out among selected groups have also shown HIV testing is more frequently reported by high-risk individuals (ie, reporting at least one risky sexual behavior).6,8,19 However, European studies have found that between 49.0% and 89.3% of individuals reporting risk behaviors had never sought voluntary HIV testing.20 It is a missed opportunity because many studies have shown that people tend to reduce sexual risk behaviors after knowing they are HIV-infected.21
Among HIV/STD-related variables, it is important to underscore self-perception of HIV risk. Studies have shown that those who perceive themselves at risk of HIV infection are more likely to be HIV tested than those who perceive themselves at low risk.19 In our study, not knowing one's risk showed lower odds of being HIV tested. In contrast, a British population-based study found that men and women who did not know whether they were at risk of HIV infection were more likely to have had an HIV test as compared with those who reported they were not at risk at all.19 This could be partially explained by referral bias among patients receiving psychiatric care, ie, a tendency to refer for testing only those who report risk behaviors assuming that those with no risk behaviors do not need to be tested.22
HIV testing entails a reasoned decision-making process involving both the individual and their healthcare providers. The majority of studies have shown the main reason to have ever been tested was associated with blood donation in men and pregnancy in women.8 We did not have data on motivational reasons to get HIV-tested. In general, adults with severe mental illness are more likely to get tested because of insurance, employment, military, immigration, pregnancy, or hospitalization as well as sexual assault or a medical reason such as illness, overdose, and blood borne infection.8,23
Our results showed that patients with higher schooling were more likely to have been HIV tested. According to França-Junior et al,14 socioeconomic inequalities persist in access to HIV testing in Brazil. Sociodemographic associations with lifetime HIV tested highlight the potential impact of poverty and illiteracy on HIV epidemic in Brazil, both affecting the implementation of prevention policies in vulnerable populations.14
We also found that participants who had above average HIV/AIDS knowledge were more likely to get HIV-tested. It is one of the first studies to provide such an important finding. In a recent study about HIV/AIDS knowledge in psychiatric patients, we found that almost 27.0% had low HIV/AIDS knowledge.13 It suggests that more targeted messages and different ways of delivering information are required to increase HIV prevention and testing.13
With regard to psychiatric conditions, patients with substance abuse were more likely to have been HIV-tested. This may actually reflect increased risk behaviors in this subgroup of patients and, consequently, higher testing rates or referral bias as mentioned before.6,8,15,16
Patients reporting other medical conditions were more likely to have been tested for HIV in their lifetime, suggesting that patients were more likely to be tested when they had access to other health services or had other medical conditions. This highlights the role of mental health services in increasing HIV testing in this population and a need to integrate AIDS and mental health care services. Similar results were found in US psychiatric patients. Those using more healthcare services and those who had recently used other health services15 were more likely to have been tested. Because mental health patients need regular follow-up, either as inpatients or outpatients, psychiatric services should be regarded as a target site for offering HIV testing on a routine basis.22 However, the present study indicates that patients who were treated only in psychiatric services are not tested at the same rate as patients who have access to more than one service.
The present study is the first one to address the issue of lifetime HIV testing in a representative sample of psychiatric patients receiving public mental health care in Brazil. However, our results may not be generalizable to all psychiatric patients because of the exclusion of more severely ill patients and those not capable of participating. In addition, the cross-sectional design limits our capacity to establish definite cause-effect relationships. Nevertheless, our findings are of concern and emphasize the need to increase HIV testing and awareness of HIV status among chronic mentally ill patients. In Brazil, most health services do not offer provider-initiated HIV testing on a routine basis, except in specific program such as HIV vertical transmission prevention strategy among pregnant women.24 In contrast, in the Centers for Disease Control and Prevention opt-out strategy, all individuals attending specified settings are offered HIV test as part of routine care but they have the option to refuse testing.25
In conclusion, this study indicates that routine HIV testing is not yet a standard practice in most mental health settings and that low HIV testing in this population is a missed opportunity for early diagnosis and treatment as well as for prevention and risk reduction. It also points out to the lack of adequate integration between AIDS and mental health services and the need to develop integral care to all patients. Although mental illness and HIV stigma may be major barriers to full access to HIV testing, mental health providers must be encouraged to play a more active role in increasing the number of patients tested. Psychiatric patients are an important vulnerable group and novel strategies for implementing more effective HIV/STD interventions for this population should be developed.
The authors thank the Brazilian Ministry of Health Department of STD, AIDS and Viral Hepatitis for supporting one of the authors to participate in a workshop during the manuscript preparation and special thanks to Ivan França Júnior for his valuable comments that helped improve the manuscript.
1. Centers for Disease Control and Prevention (CDC). Persons tested for HIV—United States, 2006. MMWR Morb Mortal Wkly Rep
2. British HIV Association, British Association of Sexual Health and HIV, British Infection Society. UK National Guidelines for HIV Testing 2008
. London: BHIVA; 2008.
3. Fernandes JR, Acurcio FA, Campos LN, et al. Initiation of antiretroviral therapy in HIV-infected patients with severe immunodeficiency in Belo Horizonte, Minas Gerais State, Brazil. Cad Saude Publica
4. Campos LN, Guimarães MDC, Carmo RA, et al. HIV, syphilis and hepatitis B and C prevalence among patients with mental illness: a review of the literature. Cad Saude Publica
5. Guimarães MDC, Campos LN, Melo APS, et al. Prevalence of HIV, syphilis, hepatitis B and C among adults with chronic mental illness: a national multicenter study in Brazil. Braz J Psychiat
6. Meade CS, Sikkema KJ. HIV risk behavior among adults with severe mental illness: a systematic review. Clinical Psychological Review
7. Guimarães MDC, McKinnon K, Campos LN, et al. HIV risk behavior of psychiatric patients with mental illness: a sample of Brazilian patients. Braz J Psychiat
8. Senn TE, Carey PC. HIV testing among individuals with a severe mental illness: review, suggestions for research, and clinical implications. Psychol Med
9. Knox MD, Boaz TL, Friedrich MA, et al. HIV risk factors for persons with serious mental illness. Community Ment Health J
10. Wainberg M, McKinnon K, Elkington K, et al. HIV risk behaviors among outpatients with severe mental illness in Rio de Janeiro, Brazil. World Psychiatry
11. Melo APS, Acúrcio FA, Cherchiglia ML, et al. Evaluation of mental health services: care and prevention of sexually transmitted diseases within PESSOAS(*) Project context [in Spanish]. Revista Medica Minas Gerais
. 2007;17(suppl 4):S240-S248.
12. Guimaraes MDC, Oliveira HN, Campos LN, et al. Reliability and validity of a questionnaire on vulnerability to sexually transmitted infections among adults with chronic mental illness—PESSOAS Project. Braz J Psychiat
13. Melo APS, César CC, Acurcio FA, et al. Individual and treatment setting predictors of HIV/AIDS knowledge among psychiatric patients and their implications in a national multisite study in Brazil. Community Ment Health J
14. França Junior I, Calazans G, Zucchi EM. Changes in HIV testing in Brazil between 1998 and 2005. Revista de Saúde Pública
15. Desai MM, Rosenheck RA, Desai RA. Prevalence and correlates of human immunodeficiency virus testing and posttest counseling among outpatients with serious mental illness. J Nerv Ment Dis
16. Goldberg RW, Himelhoch S, Kreyenbuhl J, et al. Predictors of HIV and hepatitis testing and related service utilization among individuals with serious mental illness. Psychosomatics
17. Janssen RS, Holtgrave DR, Valdiserri RO, et al. The serostatus approach to fighting the HIV epidemic: prevention strategies for infected individuals. Am J Public Health
18. Ministério da Saúde. Coordenação Nacional de DST e Aids. Fique Sabendo. 2003. Available at: www.aids.gov.br/fiquesabendo/
. Accessed May 10, 2010.
19. McGarrigle CA, Mercer CH, Fenton KA, et al. Investigating the relationship between HIV testing and risk behaviour in Britain: national survey of sexual attitudes and lifestyles 2000. AIDS
20. Renzi C, Zantedeschi E, Signorelli C; NEM Group. Voluntary HIV testing in Europe. Scand J Public Health
21. Kamb ML, Fishbein M, Douglas JM Jr, et al. Efficacy of risk-reduction counseling to prevent HIV and sexually transmitted diseases: a randomized controlled trial. JAMA
22. Satriano J, Mckinnon K, Adoff S. HIV service provision for people with severe mental illness in outpatient mental health care settings in New York. Journal of Prevention & Intervention in the Community
23. Blumberg SJ, Dickey WC. Prevalence of HIV risk behaviors, risk perceptions, and testing among US adults with mental disorders. J Acquir Immune Defic Syndr
25. Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Morb Mortal Wkly Rep
Keywords:© 2011 Lippincott Williams & Wilkins, Inc.
HIV; AIDS; HIV testing; psychiatric patients; mental health services; sexually transmitted disease