Psychosomatics 48:31-37, February 2007
doi: 10.1176/appi.psy.48.1.31
© 2007 Academy of Psychosomatic Medicine
Prevalence of HIV Infection in a General Psychiatric Outpatient Population
John L. Beyer, M.D.,
Laura Taylor, Ph.D.,
Kenneth R. Gersing, M.D., and
K. Ranga R. Krishnan, M.D.
Received November 9, 2005; revised January 11, 2006; accepted January 20, 2006. From the Duke University Medical Center, Dept. of Psychiatry, Durham, NC. Send correspondence and reprint requests to Dr. Beyer, Dept. of Psychiatry, Duke University Medical Center, Durham, NC. e-mail: beyer001{at}mc.duke.edu
|
ABSTRACT
|
The prevalence of human immunodeficiency virus (HIV) infection in the general psychiatric population is unknown. The authors conducted a retrospective review of all patients evaluated through the psychiatric outpatient clinics at Duke University Medical Center from 2001 to 2004 in order to determine the prevalence of comorbid HIV infection and mental illness. HIV infection was present in 1.2% of the psychiatric outpatients, approximately four times the occurrence of HIV infection in the general adult population of the United States. The major psychiatric diagnostic categories with a high prevalence of HIV infection were substance abuse disorders (5%), personality disorders (3.1%), bipolar disorders (2.6%), and posttraumatic stress disorder (2.1%).
|
INTRODUCTION
|
The prevalence of human immunodeficiency virus (HIV) infection among the general population of adults in the United States has been estimated to be 0.3%–0.4%.1,2 However, the risk of HIV has been found to be disproportionate among certain subpopulations. These "at-risk" groups include injection substance abusers, certain homosexual communities, urban homeless persons, and persons with severe mental illness (SMI).
In previous studies, SMI was operationally defined using a variety of criteria, but which usually included the presence of a major mental illness, chronicity of that illness, and a pervasive impairment of functioning.3,4 As such, patients with SMI are predominantly diagnosed with schizophrenia or affective disorders. They are also most frequently treated through public-sector providers or providers who accept Medicare/Medicaid.
Over the past decade, numerous articles have reported variable rates of HIV infection among persons with SMI, ranging from 3.1% to 22.9%.5–8 The high prevalence of comorbid HIV infection and SMI has prompted calls for better awareness among mental health professionals regarding the risk, diagnosis, and treatment of HIV exposure and illness among patients.9 Despite the data on HIV prevalence in SMI, there is little information about HIV prevalence among the general psychiatric population.
This retrospective study was conducted to answer three questions: 1) What is the prevalence of comorbid HIV infection and mental illness in a general outpatient psychiatric practice? 2) What is the prevalence of HIV for specific psychiatric diagnostic categories in a general outpatient psychiatric practice? and 3) How does comorbid substance abuse affect prevalence?
|
METHOD
|
Design
Using the Duke University Medical Center clinical database, we conducted a retrospective chart review of all psychiatric patients evaluated at Duke University Medical Center from 2001 through March 2004. This database contains anonymous demographic information on over 23,000 individual patients seen through the various outpatient psychiatric clinics, inpatient service, and emergency department. We limited our review only to outpatients in order to minimize crossover with the more severely mentally ill population. No individuals were included who were seen exclusively either at the emergency department or on the inpatient units. We then divided the psychiatric diagnoses into the 18 categories of psychiatric disorders identified in DSM-IV10 (Table 1) and assessed for the prevalence of HIV infection by group. Because of multiple diagnoses, a patient may have been categorized in one or more psychiatric diagnostic groups. Since mood and anxiety disorders comprised such a large class, we included a sub-analysis of each of these into the most prominent diagnostic disorders.
We then assessed the prevalence of HIV infection in all dual-diagnosis patients (psychiatric patients with a comorbid substance abuse disorder) by psychiatric diagnosis. Because all patient data analyzed in this study were anonymous, no Institutional Review Board approval was required.
Subject Sample
In all, 11,284 individuals were identified as having been provided outpatient clinical services through the department of psychiatry; 1,481 patients had a substance abuse disorder; 130 patients had HIV infection. The demographics for each of these groups are shown in Table 2.
Data Analysis
After assigning each individual to one or more of the psychiatric diagnostic groups, we calculated frequency tables for each diagnosis (Yes/No) and cross-tabulated these with HIV (Presence/Absence). By use of the SAS 8.1 program, each table was analyzed using a chi-square test (or a continuity chi-square test for tables with low frequencies; or a Fisher’s exact test for tables with very low frequencies). We also determined the proportions estimation, confidence intervals (CI) for proportions, and the relative risk (RR).
To evaluate the influence of substance abuse, we identified patients with comorbid psychiatric and substance-abuse diagnoses and created cross-tables of the diagnosis with comorbid substance abuse and presence of HIV, and the statistical tests were again performed.
|
RESULTS
|
Incidence of HIV Infection in Psychiatric Outpatients
As noted in Table 2, 130 of the 11,284 individual subjects seen through the psychiatry outpatient clinics reported AIDS or HIV infection. This is 1.2% of the total outpatient population, significantly higher (Z=8.78; p<0.001) than the national average of 0.3%.
Incidence of HIV Infection by Comorbid Psychiatric Diagnosis
Table 3 shows the prevalence of HIV-infected patients by each diagnostic category. Figure 1 graphically represents the prevalence rate of HIV for each psychiatric category. The categories that had significantly higher HIV infection rates than even the elevated rate for the Duke psychiatric sample were the following: 5%, substance abuse disorders (Z=33.05; p<0.0001); 1.4%, unipolar depressive disorders (Z=15.15; p<0.0001); 2.6%, bipolar disorders (Z=14.89; p<0.0001); 2.1%, posttraumatic stress disorder (Z=7.89; p<0.0001), and 3.1%, personality disorders (Z=15.39; p<0.0001).
View larger version (28K):
[in this window]
[in a new window]
|
FIGURE 1. Prevalence of HIV Infection in a Psychiatric Outpatient Clinic
Horizontal bar represents the mean level of HIV prevalence (0.003). See Table 1 for psychiatric disorder categories from DSM-IV-TR.
|
Incidence of HIV Infection in Dual-Diagnosis Subjects (by Psychiatric Comorbidity)
Because substance abuse has consistently been associated with HIV infection independent of mental illness, we assessed the impact that substance abuse had on the incidence of HIV infection in psychiatric patients. Of the 11,284 subjects seen through the psychiatry outpatient clinics, 1,481 had a substance-abuse disorder. This dual-diagnosis group is shown in Table 3 by comorbid psychiatric diagnosis and is graphically represented in Figure 2.
View larger version (25K):
[in this window]
[in a new window]
|
FIGURE 2. Prevalence of HIV in Dual-Diagnosis (Substance Abuse) Outpatients
Horizontal bar represents the mean level of HIV prevalence (0.003). See Table 1 for psychiatric disorder categories from DSM-IV-TR.
|
We calculated the relative risk of having an HIV infection for each psychiatric diagnostic category, depending on the presence or absence of a comorbid substance abuse (Table 4). Excluding the diagnostic categories in which there were too few subjects for an accurate estimate of HIV prevalence (Mental Disorders Due to a General-Medical Condition, Somatoform Disorders, Factitious Disorders, Dissociative Disorders, Sexual and Gender Identity Disorders, Eating Disorders, Sleep Disorders, Impulse-Control Disorders, and Adjustment Disorders), the highest risk of HIV infection in noncomorbid substance abuse was for personality disorders (10.3), bipolar disorders (8.8) and posttraumatic stress disorder (PTSD) (7.0). For those with comorbid substance abuse, the highest risk was for personality disorders (31.0), depressive disorders (30.3), delirium/dementia (25.7), general anxiety disorder (GAD) (23.7), and social phobia (20.3).
|
DISCUSSION
|
The results of this study indicate that outpatients with psychiatric illnesses exhibit an increased prevalence of HIV compared with estimates for the overall United States population. The prevalence found in this study (1.2%) was approximately four times the overall estimated U.S. population prevalence (0.3%–0.4%).1,2 To our knowledge, this is the first report of HIV infection rates in psychiatric patients who are not drawn from SMI or substance-abuse populations. It is also the first report of prevalence of HIV infection by psychiatric diagnostic category.
Our study continues to support the observation that substance-abuse disorders place persons at a much higher risk of HIV infection.11,12 On average, the presence of a dual diagnosis increased the risk of HIV exposure approximately 4 to 5 times that of psychiatric patients without substance abuse problems.
We also found that those with personality disorders (p<0.001), bipolar disorder (p<0.001), PTSD (p<0.05), and depressive disorders (p<0.05) all had significantly higher risk of HIV infection than the general psychiatric population (and even greater risk than the general adult U.S. population). One might expect that HIV rates in personality and bipolar disorders would be elevated because of the impulse-control problems involved in the diseases, whereas depression and PTSD rates may be elevated because of factors involved in HIV exposure or the disease itself. Alternatively, it is also possible that these diseases may just "self-select" for comorbid infections, since the risk of substance abuse has long been noted to be high in these disorders. However, the impact of substance abuse was different in each of these categories. Patients with bipolar disorders, PTSD, and personality disorders had a much lower substance abuse/HIV risk ratio (1.7, 2.6, and 3.1, respectively) compared with the full sample. This suggests that comorbid substance abuse, although still a significant risk factor, was less important in these conditions than in other psychiatric diagnoses. The highest risk ratio (6.4) was for depressive disorders, suggesting that, in this group, substance abuse comorbidity played a much more significant role in HIV risk.
Although the prevalence rate of HIV comorbidity in our study (1.2%) markedly exceeds that found in the adult U.S. population (estimated at 0.3%–0.4%),1,2 it is much less than prevalence rates published in studies focusing on patients with SMI. One explanation may be that in our sample, SMI outpatients are still present, but are diluted in the overall outpatient population, thus diluting the prevalence of HIV infection. Although this may be true, it does not explain all the findings. For example, the patterns of infection appear to be different. Patients with psychotic disorders (such as schizophrenia and schizoaffective disorders), who are usually defined as having SMI, actually had a lower prevalence of HIV infection than the full sample. This suggests that the psychiatric outpatient population served by the Duke University Medical Center’s psychiatric clinics is significantly different from those seen in the public-sector clinics where most of the SMI studies were conducted, and the HIV rates are not just dilutions of SMI patients.
Of special interest is the comparison demographic data for psychiatric outpatients compared with the national and state averages. In the full outpatient sample, women outnumbered men 60% versus 40%. However, HIV/AIDS was more prevalent in men than in women (69% versus 31%). This pattern is consistent with HIV/AIDS national statistics (73% versus 27%)2 and North Carolina data (69% versus 31%).13
The racial background of our outpatient psychiatric sample was predominantly white (64%). However, HIV/AIDS was more predominant in blacks than whites (52% versus 41%). This pattern is consistent also with HIV/AIDS national statistics (50% versus 29%) and North Carolina data (66% versus 26%).13
Our study does have several limitations. The first is the possibility of a sampling bias. One could argue that being a tertiary-care center and an academic institution, clinic patients evaluated at Duke University may be more likely to have increased psychiatric morbidity or medical complications, thus affecting HIV prevalence. However, review of clinical activity shows very limited association between the HIV clinics and psychiatric clinics. Furthermore, research study participants were not included in this sample.
Another limitation is the lack of a nonpsychiatric comparison group. Thus, the actual prevalence rate for the area served by the Duke psychiatric clinics is unknown. Optimally, a comparison group matched on variables of age, gender, and location (urban/rural mix) would have been available. Since it was not, we assumed similar HIV prevalence rates with the general adult U.S. population, although this may not be an accurate comparison group. Of note, North Carolina’s HIV/AIDS prevalence rate appears to be slightly lower than the national average (0.32% versus 0.26%).1
As with any retrospective study, reporting errors may skew the results. In this study, outpatient psychiatrists were responsible for documenting the medical comorbidities and psychiatric diagnoses. It is possible that patients may not have reported their HIV status, that many of the patients did not know their HIV status, that the physician may not have asked, or that the physician may not have entered all the data. Each of these possibilities would have led to an underreporting of the HIV infection rates. Furthermore, the record of HIV status was based on patients’ self-report. Laboratory verification was not required for this data collection.
Similarly, the psychiatric diagnoses were made by trained psychiatrists, but the diagnoses were not required to meet rigid research criteria. Also, we suspect that most psychiatrists gave diagnoses to only the current or primary psychiatric diagnoses. Thus, underreporting of the psychiatric diagnoses may have occurred.
Finally, generalizability from our outpatient population is not inherent for all psychiatric outpatients. We believe that our large outpatient sample represents a population that would normally be seen in an urban, general outpatient psychiatric practice. It does not, however, represent all psychiatric outpatients or practices. We believe that the importance of these data lies in highlighting the prevalence of HIV risk in psychiatric patients, the differential risk factors in these patients, and the need for more observation and understanding of how this public health problem affects the treatment we provide to our patients. Practitioners who treat substance abuse, personality disorders, bipolar disorders, and PTSD need to be especially aware of the potentially increased risk for their patients.
|
CONCLUSIONS
|
The presence of mental illness is associated with a higher risk of HIV infection, confirming that patients with psychiatric disorders are a vulnerable population. This pattern of susceptibility appears to affect psychiatric patients in addition to those with serious mental illness. Substance-abuse patients remain at the highest risk for HIV infection, but patients with other psychiatric diagnoses are also vulnerable, especially those with personality disorders, bipolar disorder, and PTSD. The presence of a dual diagnosis (particularly in patients with these psychiatric diagnoses) substantially increases that risk of HIV infection. Clinicians should be aware that HIV infection is prevalent in all psychiatric outpatients and should consider this in their evaluations and treatment plans.
|
REFERENCES
|
- McQuillen GM, Khare M, Karon JM, et al: Update on the sero-epidemiology of human immunodeficiency virus in the United States household population: NHANES III, 1988-1994. J Acquir Immune Defic Syndr 1997; 14:355–360
- Centers for Disease Control and Prevention: HIV/AIDS Surveillance Report, 2003, Vol. 15. Atlanta, GA, U.S. Dept. of Health and Human Services, Centers for Disease Control and Prevention 2004:1-46 (also available at: http://www.cdc.gov/hiv/stats/hasrlink.html)
- Goldman HH, Gattozzi AA, Taube CA: Defining and counting the chronically mentally ill. Hosp Community Psychiatry 1981; 32:21–27[Medline]
- Schinnar A, Rothbard A, Kanter R, et al: An empirical literature review of definitions of severe and persistent mental illness. Am J Psychiatry 1990; 147:1602–1608[Abstract/Free Full Text]
- Carey MP, Weinhardt LS, Carey KB: Prevalence of infection with HIV among the seriously mentally ill: review of the research and implications for practice. Prof Psychol Res Pract 1995; 26:262–268[CrossRef]
- Cournos F, McKinnon K: HIV seroprevalence among people with severe mental illness in the United States: a critical review. Clin Psychol Rev 1997; 17:159–169
- McKinnon K, Cournos F: HIV infection linked to substance use among hospitalized patients with severe mental illness. Psychiatr Serv 1998; 49:1269[Free Full Text]
- Rosenberg SD, Goodman LA, Osher FC, et al: Prevalence of HIV, hepatitis B, and hepatitis C in people with severe mental illness. Am J Public Health 2001; 91:31–37[Abstract]
- American Psychiatric Association: Practice Guidelines for the Treatment of Patients With HIV/AIDS. Am J Psychiatry 2000; 157:1–62[Free Full Text]
- American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Vol. IV-TR. Washington, DC, American Psychiatric Association, 2000
- Dausey DJ, Desai RA: Psychiatric comorbidity and the prevalence of HIV infection in a sample of patients in treatment for substance abuse. J Nerv Ment Dis 2003; 191:10–17[CrossRef][Medline]
- Stoskopf CH, Kim YK, Glover SH: Dual diagnosis: HIV and mental illness, a population-based study. Community Ment Health J 2001; 37:469–479[CrossRef][Medline]
- North Carolina Division of Public Health: North Carolina 2003. HIV/STD Surveillance Report. NC Dept. of Health and Human Services, 2003, pp 1-51 (available at: http://www.epi.state.nc.us/epi/hiv/pdf/HIVSTD2003.pdf)
Get information about faster international access.
Privacy Policy
Copyright © 2007
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
