HIV antibody in persons screened for syphilis: prevalence in a New York City emergency room and primary care clinics

A budget impact analysis of rapid human immunodeficiency virus screening in Veterans Administration emergency departments

BACKGROUND

Human immunodeficiency virus (HIV) screening is cost-effective and recommended in populations with low disease prevalence. However, because screening is not cost-saving, its financial feasibility must be understood.

STUDY OBJECTIVES

We forecast the costs of two Emergency Department-based HIV testing programs in the Veterans Administration: 1) implementing a non-targeted screening program and providing treatment for all patients thusly identified (Rapid Testing); and 2) treating patients identified due to late-stage symptoms (Usual Care); to determine which program was the most financially feasible.

METHODS

Using a dynamic decision-analysis model, we estimated the financial impact of each program over a 7-year period. Costs were driven by patient disease-severity at diagnosis, measured by CD4+ category, and the proportion of patients in each disease-severity category. Cost per CD4+ category was modeled from chart review and database analysis of treatment-naïve HIV-positive patients. Distributions of CD4+ counts differed in patients across the Rapid Testing and Usual Care arms.

RESULTS

A non-targeted Rapid Testing program was not significantly more costly than Usual Care. Although Rapid Testing had substantial screening costs, they were offset by lower inpatient expenses associated with earlier identification of disease. Assuming an HIV prevalence of 1% and 80% test acceptance, the cost of Rapid Testing was $1,418,088, vs. $1,320,338 for Usual Care (p=0.5854). Results support implementation of non-targeted rapid HIV screening in integrated systems.

CONCLUSIONS

This analysis adds a new component of support for HIV screening by demonstrating that rapid, non-targeted testing does not cost significantly more than a diagnostic testing approach.

An educational HIV pretest counseling video program for off-hours testing in the emergency department

STUDY OBJECTIVE

Multiple barriers to traditional pretest HIV counseling make HIV testing difficult to accomplish in the emergency department setting in off hours. This study compares the educational effectiveness of a 10-minute pretest counseling video with the usual practice of a session with an HIV counselor.

METHODS

This was a prospective randomized controlled trial of adult patients presenting to the urgent care area of a busy inner-city hospital. Patients either viewed an HIV educational video or spoke with an HIV counselor for pretest counseling. The video was developed by 2 of the investigators (YC, MH) and covered essential educational elements for HIV testing, as required by the New York State Department of Health. All participants completed a measure of HIV knowledge after their intervention. An equivalence analysis was performed to assess whether the video was at least as good as counseling in terms of overall mean knowledge score.

RESULTS

Of 129 patients recruited for the study, 65 patients were randomized to the intervention and 64 patients to the control group. Five patients were unable to complete the study. The final analysis was based on 124 patients. Mean knowledge scores were higher in the intervention (85.3% versus 79.7%; 90% confidence interval for the difference 2.6% to 8.7%).

CONCLUSION

We conclude that the use of an educational video with an inner-city adult population was at least as effective as in-person pretest counseling in conveying information related to HIV testing.

HIV prevalence in patients with syphilis, United States

BACKGROUND

Among persons with a sexually transmitted disease (STD), the proportion who are also infected with HIV is a major factor influencing the public health impact of that STD on HIV transmission.

GOAL

To assess HIV infection in persons with syphilis in the United States.

STUDY DESIGN

A systematic literature review was conducted of U.S. studies with HIV seroprevalence data in patients with syphilis.

RESULTS

Thirty studies were identified and analyzed. The median HIV seroprevalence in men and women infected with syphilis was 15.7% (interquartile range [IQR]: 13.6-21.8%), among men was 27.5% (23.1-29.6%), and among women was 12.4% (8.3-20.5%). Median odds ratios for men and women, men only, and women only were 4.5, 8.5, and 3.3, respectively. Seroprevalences among men who have sex with men (MSM) and injecting drug users (IDU) ranged from 64.3-90.0% and 22.5-70.6%, respectively.

CONCLUSIONS

Despite substantial variability, HIV seroprevalence is high among patients with syphilis in the United States, identifying them as a critical target group for HIV prevention efforts.

Patterns of HIV testing in Scotland: a general practitioner perspective

General practitioners are one of the largest groups who offer HIV testing but little is known about the patient group who present for HIV counselling and testing in primary care. This study describes the risk factors, positivity rate, temporal trends and demographic profile of patients presenting to their general practitioner for HIV testing in central Scotland. Of 8,466 tests taken 1% (84%) were positive with an independent association between a positive result and age group, history of drug misuse, homosexuality and region of testing. The majority of tests were performed in those who were at low risk and because of patient concern about HIV. No increase in the number of positive tests was apparent over the five-year study period. Variations between GP testing for HIV occurs in different regions and may reflect the underlying HIV positivity rate. General practitioners performing HIV tests are well placed to educate their patients about HIV and encourage low risk sexual behaviour patterns.

HIV seroprevalence in emergency department patients: Portland, Oregon, 1988-1991

OBJECTIVES

In Portland, OR: 1) to determine the changes in HIV seroprevalence for ED patients from 1988 to 1991, 2) to define the characteristics of the HIV-positive ED patient, 3) to determine the hepatitis B seroprevalence of HIV-seropositive ED patients, and 4) to demonstrate the feasibility of an ED population-based surveillance investigation.

METHODS

A prospective, multiyear observational, cross-sectional, multicenter, population-based seroprevalence study was performed using seven urban hospital EDs. Serologic testing for HIV and hepatitis B was performed on excess blood obtained from ED patients. Four sampling periods were used at each hospital at 14-month intervals starting June 1988 and ending December 1991. The blood specimens were obtained concurrently at all the participating hospitals.

RESULTS

Of 1,681 patients, 17 (1.0%) were HIV-positive. The HIV seroprevalence rate was relatively stable over time: 0.5% (2/444) in 1988, 1.7% (7/396) in 1989, 1% (3/296) in 1990, and 0.9% (5/545) in 1991. Most (94%) HIV patients were men, 100% were white, 81% were > or = 30 years old. Most (59%) of the HIV-positive patients also were positive for hepatitis B core antibody. Many (76%) of the HIV-positive patients were known to be positive by the emergency health care worker.

CONCLUSION

HIV seroprevalence among the ED patients in Portland, OR, was generally stable from 1988 to 1991. Many HIV-positive patients also were hepatitis B-positive, thus representing a double occupational infectious disease risk to ED personnel. A significant minority (24%) of the HIV-positive patients were not known to be HIV-positive by the ED personnel. Universal precautions and hepatitis B immunization are paramount for reducing the risk of infectious disease due to exposure to body fluids.

Screening and empiric treatment for syphilis in an inner-city emergency department

OBJECTIVE

To examine targeted screening and empiric treatment for syphilis in an urban ED.

METHODS

Screening of emergency patients during previously arranged shifts from July 1991 through January 1992 in a university-affiliated, inner-city ED. Emergency patients who perceived that they had high-risk factors for syphilis (i.e., cocaine or heroin use or sexual contact with a user of these substances) were compared with emergency patients denying high risk. All presumed high-risk patients and alternate patients in the group who denied high risk (control group) were screened in the ED with the rapid plasma reagin (RPR) test. Empiric antibiotic treatment was initiated if the patient was RPR-positive and gave no previous history of syphilis. In addition, serum was submitted to the state laboratory for VDRL and microhemagglutination-Treponema pallidum (MHA-TP) testing. Blinded serologic testing for HIV antibody was performed later on frozen serum.

RESULTS

Of 806 patients presenting to the ED, 276 (34%) admitted to high-risk behavior. Of 373 patients tested by RPR in the ED (216 high-risk and 157 control patients), no significant difference was found between the high-risk and the control patients in untreated syphilis [8 (4%) vs 4 (3%)] or positive MHA-TP [47 (22%) vs 25 (16%)]. In the high-risk group, the women were more likely than the men to be MHA-TP-positive (OR = 2.58, 95% CI 1.12-7.98, p = 0.04). Among the women, the MHA-TP was more often positive for the high-risk than for the control patients (34% vs 15%, OR = 2.27, 95% CI 1.12-4.67, p = 0.023). For the high-risk group, seven (3%) new cases of syphilis were managed empirically, vs three (2%) new cases for the control group. HIV antibodies were detected in 16 of 212 (8%) high-risk patients and five of 155 (3%) control subjects (p = 0.13).

CONCLUSION

This inner-city ED population has a high frequency of positive syphilis and HIV serologies, regardless of acknowledged drug use risk factors. Therefore, in areas reporting high syphilis infection rates, consideration should be given to offering screening for syphilis to all emergency patients, along with establishment of adequate counseling and follow-up.

Syphilis and HIV infection among displaced pregnant women in rural Mozambique

A cross-sectional study was conducted among displaced pregnant women in Mozambique to determine the prevalence and correlates of HIV infection and syphilis. Between September 1992 and February 1993, 1728 consecutive antenatal attendees of 14 rural clinics in Zambézia were interviewed, examined, and tested for HIV and syphilis antibodies. The seroprevalence of syphilis and HIV were 12.2% and 2.9%, respectively. Reported sexual abuse was frequent (8.4%) but sex for money was uncommon. A positive MHA-TP result was significantly associated with unmarried status, history of past STD, HIV infection, and current genital ulcers, vaginal discharge, or genital warts. Significant correlates of HIV seropositivity included anal intercourse, history of past STD, and syphilis. In summary, displaced pregnant women had a high prevalence of syphilis but a relatively low HIV seroprevalence suggesting recent introduction of HIV infection in this area or slow spread of the epidemic. A syphilis screening and treatment programme is warranted to prevent perinatal transmission and to reduce the incidence of chancres as a cofactor for HIV transmission.

Prevalence of HIV infection in oral and maxillofacial surgery patients

To date there have been no systematic reports of the prevalence of HIV infection in dental surgery patients. The purpose of this study was to estimate the prevalence of (1) HIV infection and (2) the risk factors for HIV infection in a sample of patients who went an oral and maxillofacial surgery outpatient clinic in an urban hospital setting. Using a cross-sectional study design and anonymous self-administered patient questionnaires, we estimated the prevalence of HIV infection to be 4.8% (95% confidence interval = 2.3% to 9.7%) in our sample of 165 patients. Prevalence estimates for HIV risk factors were: (1) homosexual/bisexual preference (17.3%), (2) history of sexually transmitted disease (20.0%), (3) history of intravenous drug abuse (10.0%), and (4) history of blood transfusion (3.8%). In general, one might anticipate the prevalence of HIV infection in the patient sample to reflect the community prevalence of HIV infection. Consequently, the prevalence estimates of HIV infection in a sample of patients who live in San Francisco may not be applicable to the reader's local community. Given the known risk factors of HIV infection, however, a well-directed, frank patient history may be useful in identifying patients at risk for HIV infection.

HIV seroprevalence in sexually transmitted disease clients in a low-prevalence southern state. Evidence of endemic sexual transmission

We studied human immunodeficiency virus (HIV) seroprevalence and risk factors in 3052 clients attending three large public, sexually transmitted disease (STD) clinics in central North Carolina in mid-1988. Anonymous self-administered questionnaires linked to HIV serologies obtained by testing extra blood from syphilis serologies without personal identifiers showed the following characteristics of the respondents: 60% were men, 81% were black, the median age was 24 years, 5% were injecting drug users since 1978, 7% reported a history of syphilis, and 8% of men were homosexual or bisexual. HIV seropositivity was found in 76 subjects (2.5%), including 46% of the homosexual men, 25% of the bisexual men, 1.6% of the heterosexual men, and 0.6% of the women. Elevated HIV seroprevalence rates were found in subjects with a history of or seroreactivity for syphilis (HIV-positive rate of 53% in homosexual or bisexual men, 9% in heterosexual men, 3% in women) and with histories of gonorrhea (HIV-positive rate of 37% in homosexual or bisexual men, 2.6% in heterosexual men, 1% in women), and intercourse (41% in homosexual or bisexual men, 2% in women), prostitute contact (5% in heterosexual men), and sex with casual partners (2% in women). Even a state with a low incidence of acquired immunodeficiency syndrome can include subpopulations with a high HIV seroprevalence, apparently disseminated endemically in association with bacterial STDs.

HIV and AIDS among adolescents in the United States: increasing risk in the 1990s

Acquired immunodeficiency syndrome (AIDS) and human immunodeficiency virus (HIV) are growing problems among U.S. adolescents. By examining recent data on AIDS surveillance and HIV seroprevalence, surveys on teenagers' knowledge, beliefs, and behaviors related to HIV/AIDS, key treatment issues, and barriers to prevention, this manuscript reviews the problem and proposes possible ways of combating it. African American youth have the highest rates of AIDS and white youth the lowest. However, the largest number of AIDS cases overall has been recorded in white males, reflecting relatively high case rates in boys with hemophilia and in young male homosexuals. Predominant HIV risk factors for adolescents are unprotected sex and/or sharing injection drug equipment with an infected partner. Relatively high rates of HIV infection in adolescent females may indicate their greater physiological vulnerability than adult females to sexually transmitted diseases (STDs). Data from HIV seroprevalence studies suggest a substantially increased heterosexual epidemic in the 1990s, especially in large east coast cities and southeastern rural areas where drug use and/or STDs are highly prevalent. More comprehensive prevention and treatment services are needed to prevent ongoing expansion of HIV infection and AIDS in the adolescent age group.

Prevalence of neurosyphilis in human immunodeficiency virus-infected patients with latent syphilis

PURPOSE

A prospective study was done to determine the prevalence of confirmed neurosyphilis (cerebrospinal fluid [CSF] Venereal Disease Research Laboratory [VDRL]-reactive) in human immunodeficiency virus (HIV)-infected patients with latent syphilis (reactive serum rapid plasma reagin [RPR] and microhemagglutination-Treponema pallidum [MHA-TP]).

PATIENTS AND METHODS

All HIV-infected patients seen for their first visit at the Los Angeles County/University of Southern California Medical Center AIDS Clinic from June through December 1990 were screened for latent syphilis. Those with reactive serum RPRs and MHA-TPs who had not received recent (within 6 months) therapy for syphilis were offered diagnostic CSF sampling.

RESULTS

A total of 312 patients were screened, of whom 71 (22.8%) had reactive serum RPRs and MHA-TPs. Thirty-three of these patients (47%) had diagnostic CSF sampling (26 refused lumbar puncture or were lost to follow-up; 12 had had recent therapy for syphilis and thus did not have CSF sampling). Among the 33 patients who had CSF sampling, 20 (60.6%) had normal CSF profiles (white blood cell count less than 8/mm3; protein less than 0.60 g/L; glucose greater than 2.8 mmol/L) and nonreactive CSF VDRLs. Ten of the 33 patients (30.3%) had abnormal CSF profiles and nonreactive CSF VDRLs, and three of 33 (9.1%) had reactive CSF VDRLs.

CONCLUSIONS

Asymptomatic neurosyphilis was found in 9.1% of our patient population undergoing CSF sampling, giving a 1.0% prevalence of CSF VDRL-reactive neurosyphilis in the population we screened. The abnormal CSF findings may have been due to either nonreactive CSF VDRL neurosyphilis, central nervous system infection with HIV, or infection with some unrecognized agent.