Ending AIDS as a public health threat: Treatment-as-usual risk-reduction services for people with mental illness

Abstract

Objective

People with mental illness have higher HIV infection rates than the general population. Little is known about whether care systems for this population are effectively participating in global efforts to end AIDS as a public health threat. This paper examines treatment-as-usual HIV risk-reduction services within public mental health settings.

Methods

We interviewed 641 adults attending eight public psychiatric clinics in Rio de Janeiro about their prior sexual risk-reduction program participation, HIV testing history, HIV knowledge, and sexual behaviors.

Results

9% reported any HIV risk-reduction program participation; 75% reported unprotected sex. Program participants had greater HIV knowledge (p = .04) and were more likely to have HIV testing in the last three months (p = .02) compared to non-participants. Participation did not affect sexual behaviors.

Conclusions

Including people with mental illness in global efforts to end AIDS requires a greater commitment to implementing effective interventions within public mental health systems.

Global efforts to end AIDS as a public health threat are underway¹; the extent to which public psychiatric care systems are participating is unknown. People with mental illness have higher rates of HIV infection and sexual risk behaviors than the general population, even among those with good HIV knowledge².

Steps for ending AIDS include routine testing for HIV infection; preventing infection among those who test HIV-negative; and treatment to the point of viral suppression for those who test HIV-positive to prevent forward transmission¹. Psychiatric clinics are opportune settings for services tailored to people with mental illness: clinicians and program staff can provide support, encourage participation in HIV services, establish linkages between systems of care, address barriers, and allay concerns. Strong therapeutic alliance is associated with greater acceptance of HIV testing and may have similar effects on prevention and treatment uptake³. Barriers to providing these services include limited staff training and lack of funds for condoms. Patients report that clinicians neglect their sexual needs, desires for developing romantic partnerships, and complaints about the sexual side effects of medications, all of which may affect HIV risk ². Medical care providers are more likely to report delivering HIV-related services than mental health care providers⁴.

The Brazilian government, a leader in responding to the AIDS epidemic, funded HIV educational training in the public psychiatric care system⁵. This is the first report of outcomes of HIV risk-reduction services provided in real-world public mental health programs. We examined client-level characteristics associated with having received treatment-as-usual HIV risk-reduction programming in public psychiatric clinics in Rio de Janeiro and the effects of program participation on HIV-related behaviors.

Methods

Adults self-referred or were referred by clinic providers at eight public mental health outpatient psychiatric clinics between June 2007 and November 2009 as part of an NIMH-funded study approved by both the US and Brazilian institutional review boards. A research team psychiatrist and a clinic provider evaluated patients’ capacity to consent to study participation.

Formative work preceded tailoring instruments to enhance cultural specificity ^6,7. Psychiatric diagnoses were determined using the Mini International Neuropsychiatric Interview – PLUS (MINI PLUS), developed and validated for DSM-IV and ICD-10 diagnoses with both US and Brazilian patients ⁸. Sexual risk behaviors in the past three months were assessed using the Brazil SERBAS (SERBAS-B), a semi-structured interview adapted to the local population with reasonable to excellent test re-test reliability ⁹ that asseses sexual practices, number and type of partners, condom use, whether sex was exchanged for something (e.g., money, drugs, shelter), and alcohol and other drug (AOD) use during sex. Sexual risk protective behaviors assessed for the past three months included reducing the number of sex occasions, decreasing the number of sex partners, changing risky sexual practices, and using condoms more frequently. HIV transmission and prevention knowledge was assessed using the Brazilian adaptation of the Brief HIV Knowledge Questionnaire (Brief HIV-KQ), a 17-item true/false scale, with higher scores indicating greater HIV-related knowledge (range: 0 to 17) ⁹. Participants were asked if they had undergone HIV testing in the past 3 months, and whether, within the last year, they had participated in any type of program specifically intended to help them decrease sexual risks or increase safer sex.

Descriptive statistics included range, mean, and standard deviation (SD) for continuous variables and count and percent (%) for nominal variables. Student’s t test was used for continuous dependent variables, and cross tabulation and chi-square statistics were used for nominal variables to compare participants with and without prior exposure to risk-reduction programs. Adjusted standardized residuals were computed to identify in which cell of the cross-tabulation there were significantly more participants than expected (positive standardized adjusted residual ≥ +2) or fewer participants than expected (negative standardized adjusted residual ≤ −2). For the 2 by 2 cross-tabulation, the chi-square statistic with continuity correction was used. Comparisons with p value lower than or equal to .05 were considered statistically significant.

Results

We recruited 641 outpatients who met eligibility criteria (i.e., having vaginal or anal sex in the last three months; 18–80 years old; not acutely suicidal or psychotic; primary diagnosis not alcohol/drug use disorder or developmental disability): 211 (33%) individuals had schizophrenia; 139 (22%) had bipolar disorder; 130 (20%) had non-psychotic depression; 66 (10%) had depression with psychosis; 40 (6%) had anxiety disorders; 31 (5%) had psychosis not-otherwise-specified; 20 (3%) had schizoaffective disorder; four had other diagnoses; and nine had secondary substance use disorder. Table 1 shows demographics and associations with prior sexual risk-reduction program participation.

Table 1.

Demographics and results^a.

	Prior risk reduction program (N=58)		No prior risk reduction program (N=583)		Total (N=641)		p value
	N	%	N	%	N	%
Age							0.101
Under 40	29	11	227	89	256	40
40 or Older	29	7	356	93	385	60
Gender							0.924
Men	24	9	245	91	269	42
Women	34	9	338	91	372	58
Ethnicity							0.001
White	12	6b	198	94c	210	33
Black	22	18c	103	82b	125	19
Multiracial	24	8	282	92	306	48
Education							0.171
Grade School	20	7	255	93	275	43
High School and beyond	38	10	327	90	365	57
Marital status							0.814
Married	28	9	272	91	300	47
Not Married	30	9	311	91	341	53
Psychiatric diagnosis							0.349
Severe Mental Illness	44	10	408	90	452	71
Non-Severe Mental Illness	14	8	175	92	189	29
Unprotected sex							0.828
Yes	42	9	436	91	478	75
No	13	8	145	92	158	25
Decreased number of sexual partners							0.579
Yes	7	11	57	89	64	10
No	51	9	526	91	577	90
Changed sexual behaviors							0.287
Yes	2	18	9	82	11	2
No	56	9	574	91	630	98
Used more condoms							0.176
Yes	29	11	238	89	267	42
No	29	8	345	92	374	58
HIV test							0.024
Yes	14	15c	77	85b	91	14
No	44	8b	504	92c	548	86
HIV Knowledge scored (M±SD)	10.6±1.9		10±2.5		10.1±2.4		0.036

^aNot all participants answered every question; results reflect past 3 months behaviors

^bNegative adjusted standardized residual > |2|

^cPositive adjusted standardized residual > |2|

Note: Adjusted standardized residuals = standardized difference between observed count and expected count. |2| is the threshold: if a residual is lower than 2 it is not statistically significant.

^dBrief HIV Knowledge Questionnaire (Brief HIV-KQ; 17-item),⁹ higher scores indicating greater HIV-related knowledge.

Ethnicity (those who self-identified as black or multiracial), HIV knowledge (higher scores), and HIV testing in the past 3 months were significantly associated with risk-reduction participation. No significant associations were found for sexual risk or protective behaviors or other variables.

Discussion

In this first study of the effects of participation in treatment-as-usual sexual risk-reduction programs by sexually active people within public outpatient psychiatric clinics, only 9% reported participating in any sexual risk-reduction program. This rate is dismaying given the considerable progress in understanding prevention, care, and treatment of HIV for this population. With Brazil’s comprehensive commitment to HIV prevention, including for those with mental illness, we expected much higher participation rates. Sexuality is considered a human right in Brazil, and mental health providers view helping patients to develop healthy intimate relationships as a desirable goal⁷. However, psychiatric settings in Brazil appear to have been as slow to address the HIV prevention needs of their sexually active patients as settings elsewhere⁷. This study demonstrates that treatment as usual is not sufficient, and mental health care systems need to scale up evidence-based programs and contribute to global efforts to stem new HIV infections.

Consistent with well-established theories¹⁰, risk-reduction program participation was associated with greater HIV knowledge but not with safer behavior. One-quarter of patients reported consistent condom use in the prior three months, similar to US studies ¹¹, but this behavior was not linked to participation in risk-reduction programs. Promoting condom use requires more rigorous dissemination of evidence-based interventions within the psychiatric care system.

Participating in sexual risk-reduction programs was associated with recent HIV testing; even so only 14% of risk-reduction program participants in our study reported HIV testing in the past three months. A Brazilian national multicenter study found a rate of 27% of psychiatric patients reporting ever being tested for HIV¹². Both of these rates are low for a population at elevated risk of HIV infection. HIV testing identified two people in need of HIV treatment, and diagnosis of infection is the first step in the HIV continuum of care that is at the heart of initiatives to end AIDS.

To protect against HIV infection, 10% of all participants reported decreasing their number of sexual partners in the past three months; 2% reported changing types of sexual behaviors; and 42% had used more condoms. None of these protective behaviors was associated with participation in sexual risk-reduction programs. This corroborates ample findings from the U.S. that call for implementing theory-driven, manualized, and efficacy-tested sexual risk-reduction programs to promote safer sexual behavior.

Black/multiracial ethnicity was significantly associated with sexual risk-reduction programming, in contrast with studies in other countries, where under-inclusion of minority ethnic groups in health care is a common disparity¹³. Almost half (47%) of this sample reported being married or in a long-term relationship. Future research should examine the extent to which HIV prevention is of salient concern in long-term partnerships and whether introducing condoms, a potentially difficult task, is warranted.

Relying on patient and provider referral for participation in this study and reliance on self-report are all limitations to this study. Regarding the latter, except for some protective behaviors, we used validated instruments to counter response bias and improve recall¹⁴ with documented test-retest reliability ⁹. People who participated in sexual risk-reduction programs may differ from those who did not in terms of access and reasons for participation. We obtained cross-sectional data; longitudinal studies are the next step. Caution is warranted in generalizing from the Brazil context where health care is guaranteed to countries where it is not, to adults with mental illness who are not inclined to participate in research, to those who are not in psychiatric treatment, or to those whose personal, clinical, socioeconomic, or cultural situations differ from those of this sample.

Conclusions

Almost three decades after the first descriptions of high rates of HIV infection among people with mental illness, the epidemic remains highly overrepresented and poorly addressed among this population, even where resources have been dedicated. People with mental illness are not routinely included in efforts to end AIDS. Recovery principles emphasize integrating health and mental health care for people with mental illness, but there has been limited uptake of HIV-focused health initiatives in mental health settings¹⁵. Mental health care systems must join efforts to end AIDS as a public health threat by delivering evidence-based HIV prevention services to reduce unsafe sexual behavior. This includes focusing on patient sexualilty as a component of recovery-based services and introducing newer advances such as pre-exposure prophylaxis (PrEP). Ending this epidemic will not achieved as long as public health initiatives ignore public mental health care systems.