Abstract
Dear Editor:
E
Following national recommendations by the Centers for Disease Control and Prevention, the Interim Louisiana Public Hospital (ILPH) Emergency Department implemented routine HIV testing for all persons aged 13–64 years in 2008. Although this testing was intended to identify individuals with previously unknown HIV infection, we hypothesized that some individuals that already carried a diagnosis of HIV infection may inadvertently be tested as well. As some of these retested individuals may not be engaged in care, we hypothesized that rapid HIV testing may create an opportunity to link them back into care, thus providing a possible auxiliary benefit to the routine testing policy.
We reviewed the medical records of all patients that tested positive by rapid HIV testing from February 2008 to February 2009 at the ILPH Emergency Department (ED) in New Orleans, Louisiana. The ILPH is a high-volume urban hospital in New Orleans with 60,000 ED encounters annually. 10 At the time of the study, the ED had a passive referral system for patients newly diagnosed with HIV infection. This entailed counseling patients and recommending that they establish care at the affiliated HIV outpatient clinic nearby, which holds weekly open hours for all patients referred from the hospital or ED. The clinic runs a multidisciplinary program that sees patients regardless of insurance status, including those who qualify for Free Care, a unique program for the medically indigent offered by the ILPH. The Institutional Review Board at ILPH approved this study protocol.
There were 8,204 patients screened by rapid HIV testing during the year of the study. Of those, 138 (1.7%) patients tested positive for HIV. Three (2%) were false positives, and 36 (26%) had a previous diagnosis of HIV based on medical record review. Table 1 compares the sociodemographic and medical characteristics of the newly HIV positive and previously positive individuals. Overall the characteristics are similar between the two groups, with the exception of mental illness, which was higher in the previously positive group (31%) than the newly positive group (13%). This may reflect the prior medical encounters among the previously positive group, and a higher rate of undiagnosed mental illness among the newly positive group. 11 The proportion of insured patients is also higher in the previously positive group (52%) compared to the newly positive group (23%), which is likely attributable to the former group's higher likelihood of having prior medical encounters that allowed for social workers to coordinate the acquisition of insurance.
Free Care is a program through the Interim Louisiana Public Hospital that provides care to qualifying indigent residents of Louisiana.
Mental illness was defined as any physician-documented major psychiatric diagnosis. Depression, bipolar disorder, and schizophrenia were among the most common diagnoses.
Indicates time of second rapid HIV test for previously positive patients.
Among the 36 previously positive patients, 7 (19%) were engaged care, as defined by a visit to an HIV specialist within 6 months prior to the rapid HIV test. Of the remaining 29 (81%) of previously positive patients who were not engaged in care, 13 (45%) were linked to HIV care within 3 months of the rapid test, 5 (17%) were linked within 3–12 months, and 11 (38%) were not linked to care at all over the year following the rapid HIV test (Table 2). Thus, there was 62% total linkage at 1 year for the previously HIV positive patients, compared to 60% linkage for the newly HIV positive patients. The previously positive patients linked to care in <3 months following the rapid HIV test had an average of 4.3 visits to an HIV specialist within the first year, and in the second year the average number of visits was 3.9. Eleven (61%) of the 18 previously HIV-positive patients that were linked to care initiated antiretroviral therapy within one month of their first appointment with an HIV specialist. However, 7 (39%) patients in this group were not retained in care 2 years after the rapid HIV test, defined by having fewer than 2 visits per year during the second year in care separated by 90 days. There were no significant differences from baseline in mean CD4+ cell counts or viral load measurements among the retained individuals over the 2-year period of study, although a trend towards improvement in both measurements was noted.
Retention in care is defined as less than 2 visits per year during the second year of care separated by 90 days.
This study confirms that routine rapid HIV testing in the ED can identify patients with a previous diagnosis of HIV who are not engaged in HIV care and link them back into care. The 3-month and 12-month linkage rates of 45% and 62%, respectively, are lower in this study than in other urban studies of newly diagnosed patients, which ranged from 48–90% at 3–4 months, and 64–88% by 12 months. 1,12 –18 This lower linkage rate may in part reflect the passive referral system present at the time of our study, as well as differences in population characteristics. Comparable data from rural areas is sparse, although these areas are known to face greater barriers to access and delivery of skilled HIV specialty care which may predispose to delayed care entry. 19 Also in our study, only 38% of previously HIV positive patients that were out of care were re-linked and retained in care at 2 years following a medical encounter that addressed their HIV infection. This compares poorly to other studies, in which average retention rates are 61–76% over an interval of at least 18–24 months. 12,13 These studies, however, included different populations with varied definitions of retention. 12,13 This illustrates a critical challenge in treating HIV infected individuals, whereby linkage and retention in care remained poor among disengaged patients even after they were provided opportunity to return to care.
The demographic and medical characteristics are very similar between the previously diagnosed and newly diagnosed patients. There was no difference in mean age, race, or gender variables between the two groups. The percentage of patients with CD4+ cell count <200 were similar as well, possibly suggesting a lack of consistent antiretroviral use among previously diagnosed patients since their initial diagnosis. Moreover, although 11 (61%) patients initiated antiretroviral therapy within a month of their first appointment with an HIV specialist, 3 (27%) of these patients were not retained in care. Data are lacking on the effectiveness of the test-and-treat strategy for re-engaging patients that are not naïve to antiretroviral therapy, and treatment decisions are complicated by the fact that many of these patients have adhered poorly to therapy in the past.
A passive referral system existed from the ED at the time of this study that may have adversely affected the linkage rate. Other studies have demonstrated improved linkage to care with the aid of a dedicated case manager providing assistance, and this resource has since been instituted at our study site. 15,20 The HIV specialty clinic receiving these referred patients, however, did offer a variety of ancillary services that ultimately were insufficient to retain these patients in care. These services included case management, social work, mental health and substance abuse counseling, nutrition, and a variety of physician specialists, all of which have been shown to increase retention rates in other studies. 21 Lack of insurance coverage is a less important barrier to accessing care in our population given the system of Free Care offered to qualifying low-income patients by the hospital and clinics.
The retrospective nature of this study restricted our ability to understand why the rapid HIV test was performed in those patients already known to have HIV infection. Laboratory testing is no substitute for simple medical history taking in eliciting HIV status. We also could not determine whether the positive rapid test itself had any influence over the physician's plan of care. Nevertheless, rapid HIV screening has allowed us to identify a distinct subset of HIV-infected individuals who were not engaged in care and to study their ability to re-engage in care following a defined encounter in the ED in which HIV infection was addressed. In conclusion, this study identifies a potential ancillary benefit of routine rapid HIV screening in emergency departments by re-identifying HIV-infected individuals that are not engaged in care. It also adds to the growing body of evidence calling for better interventions to engage and retain these individuals in HIV specialty care.