The Howard University Hospital Experience with Routineized HIV Screening: A Progress Report

Abstract

Background:

Howard University Hospital (HUH) is the first hospital in the nation to have instituted a hospital-wide routine rapid HIV screening campaign as recommended by the CDC for healthcare settings.

Methods:

HUH developed a protocol and implemented a hospital-wide routine HIV screening in October 2006. Rapid oral fluid-based HIV testing was conducted throughout the hospital using the OraSure OraQuick Advance Rapid HIV-1/2 Antibody Test. Patients with a preliminarily reactive test result were either referred for confirmatory testing or offered a Western Blot confirmatory test on-site and referred for follow-up care. This is a report on the progress of this program for the first eight months.

Results:

Of the 9,817 patients offered HIV testing, 5,642 consented. The mean age of the screened population was 40.7 years. Ninety percent of the patients screened were black and 55% were female. A preliminarily reactive test result was identified in 139 patients for a seroprevalence rate of 2.46%. Of these patients, 136, or 98% were black; 63% were male and 37% were female. HIV prevalence in the overall sample, among blacks, and among both black males and females peaked in the 40–54 year old age group. Challenges were experienced initially in securing confirmatory tests.

Conclusions:

Hospital-wide routine HIV screening is both possible and productive. The routine HIV screening campaign instituted at Howard University Hospital has identified a significant number of previously unidentified HIV positive persons. Success in assuring confirmatory testing and transition to care improved as time progressed.

The campaign can serve as a model for other healthcare institutions.

On September 22, 2006, The Center for Disease Control (CDC) publication, Morbidity and Mortality Weekly Report, was devoted to the presentation of “Revised Recommendations for HIV Testing of Adults, Adolescents and Pregnant Women in Health Care Settings”. This publication presented these recommendations after the CDC had undertaken an exhaustive review process aided by stakeholders at every level of involvement and concern with regards to the prevention of HIV-AIDS. In the document it was estimated that 25% of American HIV-positives, approximately 250,000 to 300,000, are unaware of their status, and are marked by higher viral loads, increased transmissibility and higher risk behavior. As a result, the number of new infections has remained relatively unchanged at 40,000 per year. Of those who test positive, 39% will carry a diagnosis of AIDS within a year (¹).

The CDC recommended that HIV screening in a healthcare setting should be routine for all patients aged 13–64 years, unless the prevalence of undiagnosed HIV infection has been documented to be below 0.1% for that setting. Verbal consent would be adequate with an opt-out policy, and prevention counseling would be optional. The assumptions were that earlier detection would decrease transmission through risk behavior modification, and decrease treatment cost by transitioning HIV-positives into care before they became symptomatic. Voluntary HIV screening should continue until it is established that the diagnostic yield is <1 per 1,000 subjects screened, at which point screening is no longer warranted (²). In populations for which HIV prevalence is low, but >0.1%, HIV screening is as cost effective as other established screening programs for chronic diseases (e.g., hypertension and colon and breast cancer) (³).

HUH is a private, urban teaching hospital located in Washington, DC, where the rate of AIDS cases is reported to be >10 times the national average: 170/100,000 for Washington, DC vs 15/100,000 for the U.S. Ethnic minorities in urban settings disproportionately share the burden of HIV infection. The ethnic breakdown of Washington, DC, according to year 2000 census data is: white 30.8%, black 60%, and Asian 2.7% and 7.9% described as Hispanic origin, race un-classified (⁴).

The inaugural HIV screening program was launched at HUH using a rapid oral HIV screening test on October 23, 2006. The screening program had two primary goals, [1] the identification of unknown cases of HIV infection and [2] the direction of identified patients to confirmatory testing and to care. The initial eight months experience is described here. In 1999, the Institute of Medicine recommended that HIV screening be offered to all pregnant women as a natural part of prenatal evaluation with the opportunity to refuse. This policy has been followed at HUH for some time (⁵).

METHOD-PROTOCOL

The protocol was developed by organizations from within HUH, the DC Department of Health (DC-DOH), with strong support from Hospital and University administrators. A core body of personnel, inclusive of postgraduate physicians, dentists and nurses were given a one hour training course on test performance with certification. Patients were excluded if they were unable to give informed consent, if they had significant mental illness, or if they had a life threatening health condition. Adolescents (aged 13–18 yrs.) required parental consent. Test kits were made available by Orasure Technologies, Inc., through the DC-DOH. The tests were conducted on three fronts: in the ED, in an allocated Outpatient HIV Testing Room, located in a high traffic area, and on the inpatient units by the trained personnel.

Subjects aged 13 years and above who received care at HUH between 10/23/06 and 6/21/07 were offered HIV screening using the Orasure OraQuick Advance© Rapid HIV-1/2 Antibody Test. Initial registration for the test was anonymous.

Registration forms in triplicate excluded name, date of birth, social security number and medical record number. A preliminarily reactive test was the indication to collect more demographic data, except in the ED, where subjects solicited were already ED registrants. One page of each form was sent to the DOH. Test results were reported to the patient or parent on-site shortly after the test was performed. All patients with preliminarily reactive results were advised to have a confirmatory test. The protocol was adjusted in the 6th month to offer an immediate Western Blot test to confirm HIV seropositivity.

RESULTS

At the end of a year, data will have been analyzed in the following manner. Primary data analysis will include the number of patients offered HIV Screening, those with a preliminarily reactive result, those who declined HIV screening, and breakdown by location of screening. Secondary analysis will include breakdown by gender, racial make-up, age group, and area of residence. The following data can be reported for the first eight months of the program:

• 9,810 Patients were offered testing, of which 5,637 consented during the first 8 months.

• 90% of consenting patients were black and 55% were female.

• A preliminary test was reactive in 139 patients (2.46%): 87 (63%) were male and 52 (37%) were female: 136 (98%) were black. 107 patients were confirmed as DC blacks.

• Black patients who were preliminarily reactive were more likely to be age > 40 years.

• Of the 9,817 subjects offered HIV screening, 4,173 refused testing. Of these, 55% were female and 40% were age 40 to 64 years.

Significant problems were experienced with securing confirmatory tests, which in retrospect should have been anticipated. Of the 130 preliminarily reactive patients identified between 10/20/06 and 4/26/07, only 39 presented at any location for a confirmatory test. Full demographic data were collected for every subject with a preliminarily reactive result. Reconnecting with patients was difficult because of patient denial and avoidance, homelessness and demographic misinformation. Follow-up was then left to the DOH. Post 4/26/07, a confirmatory test was available on site immediately following the reactive preliminary test. The initial confirmatory screening test was a Western Blot blood test, which required a phlebotomist, who was not always available. After May 20, 2007, an oral Western Blot confirmatory test became available. This test could be performed by any staff person. Since then, all preliminarily reactive patients have had an immediate confirmatory test. Of the 17 patients who tested preliminarily reactive after April 26, 2007, 13 have had a confirmatory test, bringing the total confirmed to 52.

Thirty-eight of the 52 were confirmed as HIV positive, 19 of whom are being managed at HUH, and some of the others are known to be under care at another DC clinic. Efforts are still on-going between HUH staff and the DC DOH to have a confirmatory test performed on all preliminarily reactive subjects, and to transition the confirmed positives into care.

DISCUSSION

With regards to Goal 1, the identification of unknown cases of HIV infection, there has been at least partial success. One hundred thirty-nine of 5,621 subjects were identified as preliminarily reactive during the initial 8 month period, a rate of 2.46%, or 25/1000. The geographic distribution of subjects tested reflected the primary and secondary service areas of HUH as determined independently in 2006 by a consulting firm working on an unrelated marketing project. Unfortunately, more than 40% of potential subjects approached refused testing (4,173 out of 9,817). No solution to this problem has been affected.

Plans are to analyze the reasons given for refusal and the potential obstacles to participation, improve the training and preparation of the staff and clinicians who provide screening services, and better educate the community about the benefits of screening and early detection of HIV infection

With regards to Goal 2, the performance of confirmatory testing and transitioning to care, again there is only partial success. A disappointingly small proportion of patients with a preliminarily reactive test underwent confirmatory testing during the early months of the program (only 52 of 139). Far too many were lost to follow-up. We cannot document that all of the 38 confirmed positives have transitioned to care at this point. However, the immediate performance of a confirmatory test as soon as a preliminarily reactive test result was received for a patient who was still present, has at least solved the problem of confirmatory testing.

CONCLUSIONS AND RECOMMENDATIONS

Prior to the inception of an HIV screening program, there should be some local health department media campaign aimed at reducing the high number of individuals who refuse screening. The protocol must include the availability of a confirmatory Western Blot test immediately following the preliminarily reactive result. Precise methods of monitoring subjects whose confirmatory test is positive are necessary to avoid losses to follow-up before there is successful transition to care.

Although it has been demonstrated that a hospital can successfully administer and carry out a program of routineized HIV screening, it can only have an impact on the populations in its service areas. Thus any conclusions from collected data reflecting age and gender distributions would apply only to the residents of that service area. For an entire community to benefit from routineized HIV screening, a broad representation of health care settings in the community has to participate in a collaborative effort

DISCUSSION

Baum, New York: Very nice talk and obviously a very difficult thing; and it points out what the politicization of this disease has done; and in almost any other disease you can think of, certainly syphilis testing historically being one of those, you really wouldn't have to ask the patient to be able to test them; and we are still living 26 years later with the same concept that stigmatization in this disease prevents us from just testing them; and I think that we would have been so far ahead epidemiologically and public health wise had this disease not been politicized the way it has, but very good talk. Thank you.

Wilson, Baltimore and Washington, DC: Victor, the data that you've got from your study would suggest that the incidence is lower than what has generally been assumed for that population; but the fact that 40% of the people opted out leaves a huge void there, and it would suggest also that many of the people who opted out did so for a reason - that they were concerned about being positive. Any comments about that?

Scott, Washington, DC: On the contrary Don, the 38 confirmed seropositives at this point, and the numbers may go higher, out of 5637 people tested extrapolates to a rate per 100,000 of 674. The Appleseed report puts the AIDS rate in the District at 170/100,000, which is more than 10 times the national prevalence. I believe that the population which we are screening is skewed to the high positive side, and a community-wide effort would probably lower the prevalence, but it would still be high.

The high number of refusals at 40% is certainly disappointing, and that group in addition to being fearful very probably represents a group with even higher prevalence, as approximately 250 people of this group admitted to being sero-positive. We have to do a better job of convincing all comers of the benefits of testing and early detection.