Abstract
Introduction:
We implemented triage nurse rapid HIV and hepatitis C virus (HCV) screening, in parallel with physician diagnostic testing, in our urban emergency department (ED).
Methods:
A 2-month cross-sectional survey was performed to determine the proportion of patients who correctly reported being tested for HIV and HCV.
Results:
A total of 492 patients were surveyed. Fifty-one (70%) of the 73 patients who reported being HIV tested and 372 (89%) of the 419 patients who reported not being HIV tested were correct. Thirty (60%) of the 50 patients who reported being HCV tested and 416 (94%) of the 442 patients who reported not being HCV tested were correct.
Conclusion:
Although most ED patients correctly reported whether testing was performed, there were many who did not. Although ED screening programs for HIV and HCV serve as an important venue for screening, strategies to improve communication require attention.
Introduction
The US Centers for Disease Control and Prevention and the US Preventive Services Task Force recommend universal HIV and targeted hepatitis C virus (HCV) screening in health care settings. 1 –3 In April 2014, we integrated triage nurse screening into emergency department (ED) clinical operations, according to these recommendations. We utilized a laboratory-based testing protocol and native staffing to offer, perform, and disclose results. 4
Because of concerns regarding the potential impact of screening on ED length of stay, our protocol did not require patients to wait for the results of their HIV or HCV tests prior to discharge. Additionally, we did not mandate the disclosure of negative results. Some experts have expressed concerns about patient comprehension of test processes with programs such as ours, where screening is streamlined and integrated into standard ED processes. 5
The objective of this study was to determine the accuracy of patient self-reporting of whether they were tested for HIV or HCV during their ED visit. Determining whether there are differences between what patients perceive to have occurred with actual test performance is important, as discrepancies between the 2 may have important public health, ethical, and legal implications.
Methods
Study Design
We performed a cross-sectional survey study. HIV and HCV screening had been in place for 3.5 months prior to study initiation. This study was approved by the Alameda Health System’s Institutional Review Board, and all patients provided written informed consent.
Study Setting and Population
Highland Hospital, Alameda Health System, is an urban teaching hospital in Oakland, California, with an accredited 4-year emergency medicine residency program. The annual ED volume is approximately 90 000 patients per year; 45% of patients are black, 30% Hispanic, and 20% white; 44% are female; and 85% of patients have public insurance.
HIV and HCV Screening Protocol
All screening were done by the triage nurse upon patient presentation to the ED. The protocol designated routine opt-out HIV screening for patients older than 13 years and targeted opt-out HCV screening for the subset of patients born between 1945 and 1965, as well as those patients who answered affirmatively to having ever used “a needle to inject drugs.” Patients were ineligible and not offered screening if they reported prior knowledge of HIV or HCV infection or if the triage nurse determined that they were medically unable to participate due to a high-acuity medical condition or impaired mental status. Opt-out verbal consent was utilized for screening, and the triage nurse could electronically order screening tests without requiring physician involvement. In addition to screening, physicians could order rapid HIV and HCV tests when clinically indicated. Blood was obtained using standard ED procedures. Tests were processed in the hospital laboratory using the Abbott Architect analyzer (Abbott Diagnostics, Lake Forest, IL). After receipt of the samples, the laboratory estimated turnaround time was 1 to 1.5 hours.
Survey Protocol
Patients were eligible for survey administration at the time of discharge or hospital admission if they were 18 years or older, English or Spanish speaking, and medically stable. Patients were ineligible for study participation if they were on a legal psychiatric hold, if they had language, cognitive, or medical barriers, or if their ED evaluation was still underway.
Survey Content and Administration
Five trained volunteer research assistants (RAs), unaware of the study purpose, administered the surveys during the assigned 4-hour blocks from July to August 2014. During shifts, RAs reviewed the ED electronic medical record (EMR) and identified adult patients flagged for discharge or admission. They approached these patients in a systematic manner, beginning first with the lowest acuity patients flagged for discharge and finishing with admitted patients. This method was chosen as patients admitted to the hospital in our department typically board for several hours and present a much longer time interval for survey administration. Research assistants were blinded to any test results, the reason for the patient’s visit, and were not given access to the medical record, in an effort to prevent selection bias. Further study eligibility was evaluated at the bedside, and participating patients completed written consent prior to survey administration.
Surveys were directly administered and responses recorded by the RAs using Microsoft Access (version 14.0.6; Microsoft Corporation, Redmond, WA, USA) on tablet computers. Patient-specific laboratory and demographic data were downloaded from laboratory health records (Novius; Siemens Healthcare, Malvern, PA, USA) and the EMR and linked to the spreadsheet via account numbers. These data were then deidentified and were analyzed using Microsoft Excel (version 14.3.7; Microsoft Corporation, Redmond, WA, USA).
The survey instrument was developed by the principal investigator (D.A.E.W.) as a content expert in collaboration with process experts from the department of emergency medicine. The survey instrument was pilot tested for process validity on a convenience sample of 15 patients, and refinements were made based on feedback. The survey elicited demographic information; prior knowledge of HIV or HCV infection; recall of whether screening or physician diagnostic testing was offered and, if so, which components; whether patients agreed to have testing completed and, if so, which components; whether tests were performed; and whether they were notified of their test results.
Outcome Measures
The primary outcome was the proportion of patients’ correct responses regarding whether they were tested for HIV and HCV, as established by concordance of the laboratory result with the surveyed recall. Secondary outcomes included the proportion of patients tested for HIV or HCV who reported being disclosed their test results.
Data Analysis
Descriptive analyses were performed for all variables. Categorical data are reported as numbers and percentages, and continuous data are reported as means with standard deviation (SD). All statistical analyses were performed using Microsoft Excel (version 14.3.7; Microsoft Corporation) and Stata (version 13; Stata Corporation, College Station, Texas, USA).
Results
Characteristics of Study Participants
A total of 1040 patients were assessed for eligibility, 601 were deemed eligible, and 492 participated in the study (Figure 1). The mean age was 44 years (SD = 15 years), 228 (46%) were female, 248 (50%) were black, and 427 (87%) were English speaking. Patients tested for HIV were similar to those tested for HCV (Table 1). Of the 492 patients analyzed, 98 completed HIV testing (77 screening and 21 physician testing) and 56 completed HCV testing (42 screening and 14 physician testing). One physician HIV test was positive, while 7 HCV tests were positive (2 screening and 5 physician testing).
Flow diagram: eligibility and enrollment of survey cohort. *Several patients had more than one reason for ineligibility recorded.
Characteristics of the Study Population.
Abbreviations: ED, emergency department; HCV, hepatitis C virus; SD, standard deviation.
aOther category includes data classified as other and unknown.
Main Results
The proportion of patients correctly reporting whether HIV and/or HCV testing was performed is shown in Figure 2. There were 73 patients who reported being tested for HIV, of which 51 were correct (70%; 95% confidence interval [CI]: 59%-79%). There were 419 patients who reported not being tested for HIV, of which 372 were correct (89%; 95% CI: 85%-91%). There were 50 patients who reported being tested for HCV, of which 30 were correct (60%; 95% CI: 46%-73%). There were 442 patients who reported not being tested for HCV, of which 416 were correct (94%; 95% CI: 92%-96%).
Patient understanding with regard to HIV and hepatitis C virus (HCV) testing in the emergency department. Error bars represent 95% confidence intervals.
The majority of patients were not disclosed their results. Only 5 (5%) of the 98 HIV-tested patients and 4 (7%) of the 56 HCV-tested patients reported being told their results. There were also 5 patients who reported being disclosed an HIV test result and 3 patients who reported being disclosed an HCV test result when no testing was performed. The patient who tested positive for HIV was discharged from the ED prior to his result disclosure and is currently lost to follow-up. The test was ordered by a physician as part of a diagnostic evaluation, and, according to time stamp data, the test result was not available with the ED at the time of discharge. Furthermore, the patient reported being unaware that he was tested for HIV. Only 1 of the 7 HCV-positive patients was aware of his result at the time of survey administration, though 3 had results available according to the time stamp data.
Limitations
Our findings reflect the experience of a single, urban, academic ED with a combined HIV and HCV screening and may not be generalizable to other EDs. The survey instrument was not validated, some patients may have misunderstood specific questions, and data were subject to recall and response bias. Lastly, the study was designed to describe the scope of patient understanding with HIV and HCV testing by surveying a convenience sample of ED patients. As such, the methodology does not allow for statistical analysis to determine the factors associated with misunderstanding.
Discussion
During our ED nurse screening and physician diagnostic testing program for HIV and HCV infection, we identified gaps in both patient understanding and provider–patient communication. There were a substantial number of patients who either were tested without their recollection or falsely reported being tested. Additionally, patients were rarely disclosed their test results, and communication with patients who tested positive was inadequate.
Only one prior study has quantified the number of ED patients who falsely believe that they were HIV tested. In the context of physician-initiated HIV testing, Khakoo et al demonstrated that 6% of the 276 surveyed patients incorrectly reported having an HIV test performed. 6 Our study expands on these findings, as we report the scope of patient misunderstanding with both HIV and HCV testing in the context of a fully integrated screening and testing program. We found that over 30% of the surveyed ED patients incorrectly reported that they had completed HIV and/or HCV testing. It reasons that patients who incorrectly think they were tested may apply a “no news is good news” mentality and assume that their test result was negative. Such misunderstanding may influence the decision to decline future testing and promulgate high-risk behaviors.
Many factors likely influence patient misunderstanding. Patient factors, such as low health literacy, confusion with the verbal, opt-out consent process, and assumptions that HIV and HCV testing are routinely performed when blood is drawn, may be at play. System factors, such as integrating screening at triage where patients may be focused on their primary complaint, and technical missteps, such as nurses forgetting to order a test after verbal assent was obtained, may also contribute.
We also demonstrate that many ED patients incorrectly reported that they had not completed HIV and HCV testing, when testing, in fact, had occurred. Patients tested without their knowledge may present medicolegal concerns and require ethical consideration. It is our belief, however, as well as others, that patients have a certain tolerance for testing and often don’t expect explicit consent to have testing performed. 7 Additional study to support this assumption is warranted.
Our study also shows that ED patients completing HIV and HCV testing did not learn their test results. Several reasons can explain the low rate of result disclosure. For many, they were simply discharged before results were completed, making disclosure impossible. Even when results were available at the time of discharge, however, very few patients recalled being notified (data not shown). This miscommunication may be explained, in part, by the lack of a requirement for negative test disclosure combined with an ED culture focused on the disclosure of clinically relevant, positive test results. Additionally, physicians may not have an awareness of the screening tests initiated by nurses at triage and therefore may not follow up on results, especially if they are negative.
We hope that these results are not interpreted as a call to revitalize explicit consent procedures. The advances made over the past decade to destigmatize and mitigate HIV exceptionalism through routine screening, opt-out verbal consent processes, and relaxation of counseling requirements should be applauded. 5,8,9 It remains our duty, however, to find a compromise between streamlined and sustainable practices while maintaining patient autonomy and minimizing misunderstanding.
Conclusion
Although most ED patients correctly reported whether testing was performed, there were many who did not. Such misunderstanding may pose medicolegal risks, lead to false reassurances, promote high-risk behaviors, and influence future test acceptance. Although ED screening programs for HIV and HCV serve as an important venue for screening, strategies to improve communication require attention.
Footnotes
Acknowledgments
The authors would like to acknowledge the ED staff’s commitment to making HIV and HCV screening available to our community.
Authors’ Note
Results of this study were presented in a poster session at the Society for Academic Emergency Medicine in 2015. The results were presented as an oral presentation at the 2015 National Summit on HIV Diagnosis, Prevention, and Access to Care.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: HIV FOCUS Grant Program Gilead Sciences. The principal investigator (D.A.E.W.) and program coordinator (S.K.P.) received HIV FOCUS grant funding from Gilead Sciences for partial salary support and program funding. Gilead Sciences had no role in study design, results interpretation, or manuscript preparation.