Three Decades and Counting: HIV Service Provision in Outpatient Mental Health Settings

As part of an annual needs assessment for a training program funded by the Health Resources and Services Administration, in 2017 a questionnaire was sent to directors of all licensed outpatient mental health care programs in New York State seeking information about program characteristics, HIV services offered, provider training needs, and coordination of mental health and HIV care. Programs were contacted via an e-mail list of 434 program directors provided by the New York State Office of Mental Health, followed by two e-mail reminders over 3 months, resulting in 132 surveys completed online via Qualtrics. Some programs had closed or had been absorbed by larger programs, and some large programs with multiple sites and licenses responded with one survey. Eleven programs from the list were not contacted because of failed delivery (N=7), retirement of the director with no new director in place (N=3), and lack of clinical license (N=1). Thus, the response rate was 31%, consistent with online survey response rates, which tend to be lower than rates for printed surveys (14).

Of the 132 responding programs, 55% were in urban areas, 21% were suburban, and 24% were rural (Table 1). Sixty-seven percent (N=60) of programs in areas with moderate to high HIV case rates were in urban areas, 24% (N=22) were suburban, and 9% (N=8) were rural. Most programs (N=110, 83%) served over 200 clients annually; nearly all (N=116, −97%) served clients who had co-occurring substance use disorders. Most programs (66%) reported treating patients known to be HIV positive. Almost one-third (27%) either treated no patients known to be HIV positive or were unable to estimate the number of HIV-positive patients in their programs. Nearly half of programs (46%) reported that they treated one to 10 patients with HIV or AIDS, annually, and 23% served more than 100 patients with HIV or AIDS, annually.

Services provided included HIV educational materials (N=62, 60%), risk-reduction interventions (N=49, 47%), HIV test counseling (N=29, 28%), support groups for HIV-positive clients (N=20, 21%), and on-site HIV medical services (N=21, 20%). Routine HIV risk assessment at intake occurred in 60% (N=60) of programs, and 33% offered HIV testing on site when risk was revealed (Table 1). Referrals for off-site HIV testing were offered by 52% of programs, and 8% of all programs reported no procedure in place for referral to HIV testing (Table 1). Condoms were not distributed by 39% of programs (Table 1). Reasons given were “no need” (N=18, 18%), lack of funds (N=17, 17%), and clinic policy prohibiting distribution (N=12, 12%). Services corresponding to the state’s ETE plan included offering HIV testing to clients (N=35, 35%), PrEP education (N=33, 32%), and PrEP prescriptions (N=21, 20%).

Whereas 25% of programs (N=25) reported that more than 75% of their HIV clients were receiving antiretroviral therapy (ART), 6% of programs (N=6) reported that none of their HIV clients were receiving ART, and 50% of programs (N=50) were unable to estimate how many of their HIV clients were receiving ART. Patient care coordinators for HIV-positive clients were available in 28% (N=28) of programs. Integration of HIV medical services with psychiatric services for HIV-positive clients was rated from 0, none, to 4, full integration. Twenty-two percent (N=14) of programs self-rated as having no integration, and 18% (N=13) rated themselves as fully integrated. Barriers to addressing HIV-related needs were lack of training (N=40, 39%), lack of time (N=8, 8%), client resistance (N=6, 6%), and staff feeling overwhelmed by patients’ comorbid conditions (N=5, 5%).

HIV services for mental health clients were rated as either “essential” or “very important” by 69% of programs, “somewhat important” by 23%, and “not very important” or “unimportant” by 8% (Table 1). Geographic differences were found for 19 of the 30 (63%) services we examined, with more reported HIV-related service delivery in urban than in suburban and rural areas.

Chi-square analysis of all comparable variables between current responses and data from similar surveys in 1997 (5) and 2004 (7) yielded some significant differences (Table 1). Since 1997, representation of rural sites increased (from 12% to 24%), fewer programs served a patient census of under 100 (from 18% to 13%), the proportion of programs serving over 500 patients per year almost doubled (from 35% to 66%), fewer programs had no patients known to have HIV/AIDS (from 15% to 9%), and more programs served over 100 patients known to be living with HIV/AIDS (from 8% to 23%). Since 2004, proportionally more programs reported no integration of HIV medical services with psychiatric services (from N=50, 14% to N=14, 22%), or rated themselves as fully integrated (from N=46, 13% to N=12, 18%).

Epidemiological evidence indicates that clients of mental health programs are at increased risk for HIV infection, and we found that nearly 70% of program directors rated HIV-related services as either “essential” or “very important.” However, most of New York’s licensed outpatient mental health clinics are not optimized to meet the HIV-related needs of clients, even in regions of the state with moderate to high HIV case rates. Several significant findings emerged in comparing the current data with data from 1997 and 2004. More mental health programs have larger caseloads of patients known to have HIV or AIDS despite the 2017 sample comprising more rural agencies and the lack of case-finding processes (reported in 40% of programs).

We found that more programs assessed HIV risk as a routine part of the intake procedure, from 30% of programs in 2007 to 60% in 2017. Increases in HIV testing were also found over the same period. Significantly more clinics either performed their own HIV testing or referred a client for such testing. The proportion of programs reporting that they had no procedure in place for HIV testing or referrals declined from 19% to 8%. Patients with serious mental illness may continue to need HIV test counseling that takes their mental health conditions into account; such counseling is unlikely at HIV testing sites where counselors are not trained to work with people with serious mental illness.

Prevention and care services appear to have declined in outpatient mental health programs: 40% reported not providing condoms to clients who were unlikely to be able to afford them, reducing the likelihood of their use. The relatively high proportion of current programs that were unable to estimate clients’ rates of HIV testing and ART initiation portrays the lack of reach of ETE into the main system of care for this population. It is critical to continue to offer both behavioral and biomedical approaches to prevent acquisition and transmission of HIV. PrEP, which is underutilized in mental health programs and nonpsychiatric settings alike, is offered to people with serious mental illness through only one in five mental health programs we surveyed, limiting access to one of the most efficacious HIV prevention interventions.

Data were provided by directors or their designees of programs whose e-mail addresses were provided to us by the New York State Office of Mental Health. Our response rate was in the acceptable range for online surveys, although the total number of programs eligible to be surveyed had reduced considerably since prior surveys were conducted. This reduction likely reflects consolidation of programs—we found that the total number of programs decreased while the proportion of programs serving more than 500 clients had nearly doubled—a trend that is also happening in other parts of the health care environment (15). The reduction also may be the result of shifting from a paper-based to an online survey method in 2017. The acceptable response rate we obtained supports the validity of our results (16); however, it was not possible to analyze differences between respondents and nonrespondents to ascertain the representativeness of participating programs, so results should be generalized with caution. On the other hand, rural counties were better represented in the current data than in prior surveys, possibly providing a more complete snapshot of HIV service provision in rural mental health programs than prior surveys did.

Data about how outpatient mental health programs are participating in New York’s initiative to end the HIV epidemic are scarce. Our findings argue for further investigation into improved strategies for integrating services. Interagency activities between health and mental health departments could be initiated or bolstered to meet the needs of people with these co-occurring conditions, with particular attention to enhancing integrated services in suburban and rural areas and to improving surveillance and monitoring systems to illuminate which clients are accessing care in both systems.

People with serious mental illness receive some needed HIV services in public outpatient mental health programs. Integrated care models exist for providing basic medical care in mental health programs or basic care for common mental disorders in primary medical care (17). Clients with intensive medical and psychiatric needs may be unable to access a single setting that can meet their needs. New York’s ETE initiative must expand to reach patients in this population and the providers who care for them.

The authors thank Marc Manseau, M.D., and Michael Compton, M.D., for acting as liaisons to the New York State Office of Mental Health licensed clinics and for offering input on survey content.