Abstract
It has been previously shown that patients may present for multiple health issues in the years preceding their initial HIV diagnosis. This retrospective cohort study analyzed the data of patients with a new HIV diagnosis, at Ochsner Health System between January 1, 2011, and December 31, 2012. The primary end point was missed opportunities, the number of healthcare visits these patients made in the 2 years prior to being diagnosed with HIV. The 125 patients in the study cohort had 649 healthcare visits during which an HIV test was not performed. These missed opportunities are the key to capturing the undiagnosed and unaware HIV-positive individual. Primary care is an ideal setting to conduct HIV testing for those who have access to regular health care. However, nontraditional providers should also be encouraged to conduct HIV testing regardless of their ability to provide treatment because evidence shows that knowledge of the diagnosis may change behavior.
Introduction
A number of studies have shown that patients may make multiple healthcare visits in the years preceding their first HIV diagnosis. 1 –4 These so-called missed opportunities are the key to capturing the undiagnosed HIV-positive population and controlling the spread of the virus.
Current guidelines recommend routine opt-out HIV testing for individuals aged 13 to 64 years regardless of risk factors. 5 Although the 2006 US Centers for Disease Control and Prevention recommendations were well received by healthcare providers, the persistence of multiple barriers to routine HIV testing prevents real-world application. 5 Providers report that insufficient time, the laborious counseling and consent process, and inadequate resources are consistent barriers across a variety of healthcare settings. 6 Physician-perceived risk also plays an important role in the proposal of an HIV test. Heterosexual transmission, transmission later in life, and patient nondisclosure may conflict with stereotypes held by physicians. Risk-based assessments have been shown to be impractical, and the use of universal opt-out testing will continue to become more important as the HIV-infected population continues to evolve. 3,7
Data from 2000 to 2010 show the HIV infection rate in Louisiana has increased (23.2-24.4/100 000), whereas the national rate has steadily decreased (17.3-13.0/100 000). 8 At 24.4 cases per 100 000 persons, Louisiana has the second highest rate of HIV infection of all US states (Maryland is the first). And according to a 2008 US Census Bureau survey, Louisiana has the third highest percentage of uninsured individuals (after Texas and New Mexico). 9 In a community with an already significant HIV burden and so many inhabitants without health insurance, there lies a degree of uncertainty surrounding the true burden of disease and the shortcomings of the current HIV testing system in capturing the undiagnosed.
An HIV test should ideally be administered in a primary care setting, with pretest counseling, follow-up, and treatment offered. However, in a community with a high burden of disease and a significant number of people without health insurance, alternative venues must be considered. Concerns regarding linking unfunded individuals to care after an HIV diagnosis are valid. However, regardless of the ability to receive HIV treatment, knowledge of the diagnosis may change individual behavior. 1 Furthermore, only when we understand the true burden of disease on the population can we begin to propose solutions.
The purpose of this study was to determine the number of traditional and nontraditional healthcare visits during which an HIV diagnosis could potentially have been made at Ochsner Health System (OHS) facilities in New Orleans, Louisiana.
Methods
This retrospective cohort study included all patients with a new HIV diagnosis identified between January 1, 2011, and December 31, 2012, at OHS. Potential study participants were obtained from the positive HIV enzyme-linked immunosorbent assay (ELISA) tests performed at the Ochsner Medical Center Blood Bank, where HIV testing for all Ochsner facilities is performed. This source identified all newly diagnosed patients in the system. Each chart was reviewed to determine patient eligibility. Previously diagnosed HIV-positive patients occasionally undergo repeat ELISA testing for various reasons, and those patients were excluded. The electronic medical records of each eligible patient were individually reviewed. Data were obtained from the 2-year period prior to the date the ELISA test was reported as positive. Age at diagnosis, race, initial CD4 count, opportunistic infections, risk factors, and the numbers and types of healthcare visits were recorded. Demographic information collected from each patient was compared to Louisiana and to national data using the Cochran-Mantel-Haenszel test.
Primary End Point
Presentations to OHS facilities for both traditional and nontraditional healthcare visits (Table 1) in the 2 years prior to patients’ positive HIV diagnoses were identified as missed opportunities.
Definitions.
Secondary End Point
Individuals were identified as late to care if their CD4 count fell below 200 cells/mm3 or they were diagnosed with an opportunistic infection in the year immediately following their first HIV diagnosis.
Results
The Ochsner Medical Center Blood Bank logbook recorded 159 positive ELISA tests during the study period. Of those tests, 18 were found to be indeterminate by Western blot and 16 were previously diagnosed patients. The remaining 125 patients were identified as new diagnoses and included in this study.
The newly diagnosed patients at Ochsner facilities ranged in age from 17 to 79 years and were unique compared to state and national data (Table 2). The Ochsner population had significantly fewer diagnoses in the 19 age-group (Ochsner: 3.20%, LA: 7.49%, US: 5.09%, χ2 = 41.8173, P < .0001, α = .05), as well as significantly more diagnoses in the ≥65 age-group (Ochsner: 5.60%, LA: 1.40%, US: 1.92%, χ2 = 41.8173, P < .0001, α = .05). In the Ochsner population, significantly more women were diagnosed compared to the state and national data (Ochsner: 34.40%, LA: 29.17%, US: 21.00%, χ2 = 64.4999, P < .0001, α = .05). African Americans were disproportionately affected in both the Ochsner and Louisiana state populations compared to national data (Ochsner: 71.20%, LA: 73.63%, US: 47.05%, χ2 = 415.4701, P < .0001, α = .05). Significantly fewer Hispanic or Latino patients were newly diagnosed in the Ochsner and Louisiana data compared to the national data (Ochsner: 6.40%, LA: 6.32%, US: 20.28%, χ2 = 415.4701, P < .0001, α = .05). Ochsner had significantly fewer patients reporting high-risk behaviors of intravenous drug use (IDU) or men having sex with men (MSM) compared to state and national data (IDU—Ochsner: 2.40%, LA: 10.92%, US: 3.95%; MSM—Ochsner: 12.80%, LA: 52.57%, US: 49.80%, χ2 = 249.5834, P < .0001, α = .05).
New HIV Diagnoses.
Abbreviations: IDU, intravenous drug use; MSM, men having sex with men.
Collectively, this population made a total of 649 healthcare visits in the 2 years preceding their HIV diagnosis (Table 2). Primary care clinics had the greatest number of missed opportunities (218, 33.6%), followed by emergency/urgent care (183, 28.2%), and specialty clinics (123, 19.0%). Inpatient care had the fewest missed opportunities (20, 3.1%). Inpatient care made the most diagnoses (41, 32.8%), and surgical specialties made the least (4, 3.2%).
Forty-five percent of the study population were deemed “late to care” by receiving an AIDS diagnosis at the time of their initial HIV diagnosis or in the year immediately following the initial diagnosis (Table 3). Of the late to care patients, 50 were diagnosed by CD4 <200 cells/mm3 and 6 by opportunistic infection. Of those patients who were diagnosed in the inpatient setting, 50% were considered late to care. Twenty-six percent of patients with new diagnoses did not have a CD4 count recorded, so disease progression at the time of diagnosis could not be determined.
Late to Care.
Discussion
The age at first diagnosis is significantly older in the Ochsner population compared to the state and national data (P < .0001). Limited access to medical care may contribute to later diagnoses as Louisiana has been found to have a large proportion of uninsured individuals. More women were diagnosed at Ochsner compared to the state and national data, but given these small numbers, it is difficult to draw any conclusions. Ochsner data showed fewer patients reporting high-risk behaviors of IDU or MSM compared to the state and national data. A significant portion of the data were retrieved from an electronic medical record that has since been replaced with a more comprehensive system. Men having sex with men or IDU behavior data during the period of interest was limited to what was reported in individual encounter notes or listed on the main problem list. There is unlikely to be a significant difference in the population demographics but rather a lack of documentation, therefore, no conclusions can be made at this time.
The 125 patients included in the study had a total of 649 healthcare visits at which an HIV test was not performed. Although a significant proportion of these missed opportunities occurred in primary care facilities (33.6%), primary care physicians also made a significant number of diagnoses (26.4% of the total new diagnoses), resulting in 13.1% of visits to primary care facilities where an HIV test was performed in an HIV-positive individual (see Table 4).
Diagnosis by Facility.
Abbreviation: NA, not available.
Inpatient care had relatively few missed opportunities and a diagnosis rate of 67.2%. Some patients who do not receive routine health care present with advanced disease states that require hospitalization and prompt testing. Fifty percent of patients diagnosed in the inpatient setting also received an AIDS diagnosis.
The significant proportion of patients (P = .0092) with new HIV diagnoses who presented with AIDS may be reflective of limited access to health care, lack of health-care attendance, or testing not being offered to patients across all healthcare settings. More than one-fourth of our sample did not have a CD4 count, viral load, or follow-up recorded, so the progression of their disease could not be determined. This lack of documentation is at least partly due to a referral system for uninsured patients to government-funded clinics.
Traditionally, HIV testing is performed in a primary care setting where pretest counseling, follow-up, and referral to additional services can be offered. However, more than 66.4% of missed opportunities occurred at nontraditional healthcare visits: inpatient care, specialty care clinics, surgical specialties, ob-gyn visits, and emergency care facilities. These nontraditional services accounted for more than 71.2% of the new HIV diagnoses (see Table 4).
Conclusion
In this sample population, a significant proportion of new HIV diagnoses were made during nontraditional healthcare visits. Nontraditional providers should be encouraged to continue testing to decrease the undiagnosed, unaware population and minimize morbidity and mortality associated with late diagnosis. Nonprimary care providers should consider offering HIV testing, regardless of their ability to provide ongoing HIV treatment and follow-up, as there is evidence that knowledge of a diagnosis may change individual behavior and prevent the spread of disease.
Primary care can still be considered an ideal setting for HIV testing of those who have access to regular health care. However, progress needs to be made in HIV testing in primary care. Physicians should be encouraged to implement routine opt-out testing for all patients regardless of age, sex, or perceived risk factors.
Footnotes
Acknowledgment
The authors acknowledge the expertise of Dana Schmucker, MSPH, in the statistical analysis of our data.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.