The Centers for Disease Control and Prevention (CDC) estimates indicate that approximately 1–1.2 million people, or 0.5% of the general US population aged 18–49 years, are infected with HIV, and one-quarter of these are unaware of their infection [1,2]. Early diagnosis of unrecognized infection increases the opportunity to receive potential life-sustaining clinical treatment and preventive services at the individual level  and can bring about decreased HIV transmission at the community level .
For more than a decade, public health leaders have recognized emergency departments (EDs) as one of the most critical venues for the implementation of HIV testing, with the most recent recommendations suggesting that testing be routine [5–7]. Two recent findings which have propelled the ED to the forefront of the national response plan to the ongoing HIV epidemic are that EDs are the most frequent sites for encountering ‘late testers’ (i.e. patients diagnosed with AIDS within 1 year of initial HIV diagnosis) , and EDs are the most common healthcare setting where ‘missed opportunities’ for HIV testing occur . Additional support for the importance of EDs as an essential component of the public health response includes the long-standing recognition that EDs serve as the safety net (i.e. the only point of medical contact) for minority and underserved populations, and data from numerous ED-based epidemiologic studies that report high HIV seroprevalence, 2–13% [10–16], and high rates of unrecognized infection, 0.6–3% [17–22], in many inner-city settings.
The 1993 CDC recommendations suggested that acute care hospitals with a seroprevalence of at least 1% or AIDS diagnosis rate of at least 1.0 per 1000 discharge should strongly consider offering routine HIV counseling and testing (HIV CT) to patients aged 15–54 years . Revision of these guidelines in 2001 added specific reference to using EDs as one of several nontraditional sites for expanded testing . In spite of these recommendations and evidence of the effectiveness of ED-based HIV testing, studies suggest that HIV testing occurs relatively infrequently in EDs. Several survey studies which garnered information directly from providers or academic ED directors report low rates of both conventional and rapid HIV testing, even among populations with recognized risk factors for HIV, such as sexually transmitted diseases (STDs) [23–26], and even since 2001 . Although these studies provide important insights, the extent to which ED patients undergo HIV testing and the characteristics of those tested remain unknown. National data regarding HIV testing rates of patients in ED settings are reported in the National Hospital Ambulatory Medical Care Survey (NHAMCS) annual report but are limited in that only point prevalence data are provided [27–33]. We therefore analyzed the NHAMCS data over the 13-year period prior to the new expanded 2006 CDC recommendations to estimate temporal trends in HIV testing and evaluate characteristics of patients tested for HIV in US EDs.
The ED component of NHAMCS, a well established four-stage probability survey designed by the National Center for Health Statistics, samples geographically defined areas, hospitals within these areas, emergency service areas within these hospitals, and patient visits to these emergency services areas. NHAMCS is designed to collect national representative data on utilization and provision of ambulatory care services in hospital EDs, exclusive of federal, military, and veteran administration hospitals . Trained hospital staff are instructed and monitored to complete standardized patient record forms for a systematic random sample of patient visits during a randomly assigned 4-week reporting period. Data are obtained on demographic characteristics of patients, expected source(s) of payment, patients' complaints, diagnostic/screening services, procedures, medication therapy, physicians' diagnoses, disposition, and hospital characteristics. Data consistency is routinely verified on data entry and coding through computerized algorithms with very low error rate .
We defined a patient visit as having HIV testing performed in the ED if the presence of HIV serology was either ordered or provided (variable HIV serology) in the database. NHAMCS patient record forms collect up to three patient complaints per patient visit, which are coded according to the standardized patient reason-for-visit classification system . Up to eight medications are ordered or provided, and up to three corresponding National Drug Code Directory drug classes for each medication are collected and coded in the data set. We transformed the data using the two-digit National Drug Classification (NDC) system, which groups drugs into 21 major medication classes. Up to three diagnoses are recorded as free text and coded centrally by Constella Group, Inc. and subject to quality control procedures  according to the International Classification of Diseases, ninth revision, clinical modification (ICD-9-CM). We transformed the discharge diagnoses into 16 disease groups using Healthcare Cost and Utilization Project Clinical Classifications Software . A diagnosis of HIV infection was defined as ICD-9-CM codes of 042.xx, 043.xx, 044.xx, or 795.80 for years 1993–94, and code of 042 or V08 for years 1995–2005. The NHAMCS data are publicly accessible and deidentified. Accordingly, the Johns Hopkins University Medicine institutional review board (IRB) deemed our study as nonhuman-subjects research, exempt from IRB review.
NHAMCS data from the years 1993–2005 were merged for data analysis. All patients aged 13–64 years, the target age group for CDC recommendations, were included. A sample weight that considers selection probability, nonresponse adjustment, and ratio adjustment for different total sample size each year is assigned for each patient visit to generate unbiased national estimates of ED visits . Although the sampling fraction is relatively small, the weighted numbers calculated by the method suggested by the CDC, which accounts for the four-stage probability sampling scheme, represent unbiased national estimates of the US ED population. Adjusted sample weights, strata, and primary sampling unit design variables provided by NHAMCS were used in all analyses by using Statistical Analysis Software (SAS) 9.1 SURVEYFREQ and SURVEYLOGISTIC procedures (SAS Institute Inc., Cary, North Carolina, USA). Results were reported as raw numbers for individual characteristics of interest, along with corresponding weighted frequencies, percentages, and 95% confidence intervals (CI). Bivariate analyses were performed to evaluate HIV testing rates by sociodemographic characteristics as well as ‘reasons for visit’. Bivariate analysis and estimation of 95% CI was not conducted if the sample size of interest was less than 30, as the estimate is considered unreliable under NHAMCS analysis recommendations . Multivariate logistic regression was performed to assess the odds of receiving HIV testing in a patient with a particular ‘reason for visit’ with adjustment for age group (<40 years versus ≥40 years), race (black versus others), calendar year group (1993–96, 1997–99, 2000–02 versus 2003–05), type of insurance payment (private, public, self-pay, work compensation versus others/unknown). All percentages presented were weighted percentages unless indicated.
During the 13 survey years, NHAMCS collected data on 247 179 ED patient visits, corresponding to 867 million visits for those aged 13–64 years. Of these, 1059 had HIV testing performed, corresponding to weighted 2.8 million visits (95% CI, 2.4–3.2 million) or 3.2 per 1000 ED visits (95% CI, 2.8–3.7) (Table 1). Testing rates differed by age, race, ethnicity, insurance type, and geographical region but not by sex or metropolitan status of the hospital. Highest testing rates were observed in those who were 20–39 years of age, African–American, Hispanic, recipients of worker's compensation as payment type, and patients of US northeast geographic region's EDs.
Testing rates varied by calendar year (P < 0.05) (Fig. 1a and b). During the first 4-year study period, the average rate was 3.9 per 1000 visits. Rates then fell to 2.8 per 1000 visits during 1997–1999, leveled off during 2000–2002 (2.4 per 1000 visits), and increased to 3.8 per 1000 visits during 2003–2005. Testing rates in those who were 40 years of age or older, men, white, recipients of worker's compensation as payment type, and patients of US northeast and south geographic region's EDs differed by study periods (data not shown).
Among the 1059 ED visits tested for HIV, 6% (n = 69) were actually given that diagnosis during the visit in question. The majority of them were aged 30–39 years (50%), men (71%), and recipients of public medical insurance as payment type (58%). Approximately 50% of them were African–Americans (48%) and patients of US south region's EDs (49%). No specific trend of the distribution among those tested for HIV and with a given diagnosis of HIV infection was observed by study periods.
Among those tested and those not tested for HIV, the 10 most frequent reasons for visit are shown in Table 2. Among those tested, having an HIV test performed was associated with a visit for fever, puncture wound/needlestick, exposure to bodily fluids of another person, rape, or having symptoms relating to psychological/mental disorders during bivariate analysis in which adequate sample size was attained (n ≥ 30). For three variables (puncture wound/needlestick, exposure to bodily fluids of another person, or rape), the odds ratios (OR) of having HIV testing performed were 90 or greater for each [unadjusted ORs and 95% CIs, 130.8 (89.2–192.0); 314.4 (183.3–539.3); 94.4 (60.8–146.5), respectively]. Of note, the overall rate of testing for each of these groups was relatively low at 28, 49, and 24%, respectively. Further, multivariate logistic regression analyses revealed that patients who presented with fever or symptoms relating to psychological/mental disorders were more likely to be associated with testing [adjusted ORs and 95% CIs, 1.88 (1.30–2.74); 1.87 (1.18–2.96), respectively].
The leading NDC medication classes prescribed in EDs to those tested for HIV were antimicrobials (32%), drugs for pain relief (24%) and the central nervous system (12%) as compared with drugs for pain relief (44%), antimicrobials (20%), and drugs for the respiratory system (14%) to those not tested. The leading ED diagnosis among those tested was injury/poisoning (30%) followed by infectious diseases (18%), mental disorders (13%), and diseases of the respiratory system (12%) and the genitourinary system (12%).
Among those tested for HIV and given a diagnostic code of infectious diseases (n = 180), the breakdown of diagnostic coding was as follows: 34% (or as noted above, 6% of all those tested) HIV infection, 29% STDs, 11% hepatitis, 9% pneumonia, 7% opportunistic infections (oral thrush, esophageal candidiasis, Pneumocystis carinii pneumonia, and Kaposi's sarcoma), and 4% tuberculosis. Notably, the majority (69%) of those with a diagnosis of infectious disease and pneumonia had a diagnosis of HIV infection. Those with diagnoses of other infectious diseases such as STDs, hepatitis, or tuberculosis were less likely to be coded as coinfected with HIV (2, 6, and 25%, respectively).
Raw numbers in the NHAMCS revealed that 17.0% (n = 210; 95% CI, 13.6–20.3%) of those tested were admitted and 2.8% (n = 38; 95% CI, 1.5–4.2%) were transferred to other facilities as compared with 9.5% (n = 24 784; 95% CI, 9.1–9.8%, P < 0.001) and 1.8% (n = 4826; 95% CI, 1.7–1.9%, P > 0.05), respectively, among those not tested (n = 246 120). Admission to ICU/critical care unit (data available only for 1995–2005) was 2.0% (n = 20; 95% CI, not estimated) as compared with 1.0% (n = 2084; 95% CI, 1.0–1.1%) among those not tested (n = 210 565). There were no ED deaths among those tested as compared with 410 deaths (0.2%) among those not tested (n = 246 120).
To our knowledge, this is the first national-representative population-based multiyear study depicting HIV testing rate in EDs, either prior to, or since the release of the 2006 CDC guidelines that recommend routine HIV testing in all healthcare settings . Our results demonstrate that HIV testing rates in EDs were extremely low during this 13-year study period (<0.4%). Low rates of HIV testing remain even since 2001 when the CDC explicitly advocated including ED settings as a site for routine testing. This finding is further noteworthy considering long-term advocacy for at least targeted HIV CT in acute-care settings by some experts in the field of emergency medicine [5,6,13,39–41].
The CDC recommendations for HIV testing in ED settings have evolved with the HIV epidemic. In 1993, the CDC recommended confidential voluntary HIV CT in acute-care settings with consideration for adopting routine HIV CT for all patients aged 15–54 years in settings with a seroprevalence of at least 1% or AIDS diagnosis rate of at least 1.0 per 1000 discharge . The 2001 recommendations placed increased emphasis on testing in nontraditional sites (including EDs), not previously considered appropriate for HIV counseling, testing, and referral (HIV CTR) . Recommendations remained the same for those settings in which seroprevalence was at least 1%. For settings in which seroprevalence was less than 1%, the recommendations suggested testing for those with self-reported risk profiles or presence of clinical signs/symptoms or both . The latest 2006 guidelines advocate opt-out routine HIV screening for all patients (including those in EDs) aged 13–64 years, unless the documented prevalence of undiagnosed HIV of less than 0.1% and diagnostic testing for all patients with signs/symptoms consistent with HIV infection, opportunistic illness characteristic of AIDS, or suspected acute retroviral syndrome .
Despite the fact that the CDC guidelines prior to 2006 advocated both routine and risk-targeted HIV testing in EDs, our current study and three previous surveys of ED providers or directors [23,25,26] indicate a problem with acceptance or ability to carry out the recommendations in ED settings. Our study differs from previous ones [23,25,26], in that we were able to report on actual tests done (as compared with perceived tested rates reported by ED providers), but are consistent with regard to summary findings. The three surveys showed testing rates highest in those with identifiable risks, ranging from 54–69% for occupational exposure, 36% for suspected AIDS-defining illness based on clinical manifestations, approximately 10% for injection drug use, and 3–13% for those with STDs [23,25,26].
Multiple investigators have described the challenges associated with establishing the necessary infrastructure for HIV testing in EDs [40,42,43]. Reported barriers include, but are not limited to, competing medical priorities, lack of sustainable funding, time requirements, individual state counseling requirements, patient confidentiality, and challenges with follow-up for confirmatory western blot results and linkage to care [40,42]. Finally, some leaders in emergency medicine question the wisdom and burden of public health initiatives .
In our study, we found that certain demographic subgroups had significantly higher testing rates, consistent with the pattern of the US HIV epidemic over the past 15 years [45,46]. However, our data also showed that rates were similar across both sexes, different from that reported from the epidemic nationally, in which men have outnumbered women by three- to five- fold. One plausible explanation for this is that women (compared with men) ED visits for abdominal pain may have been considered a possible sign of STDs or pelvic inflammatory disease, both of which are recognized to be associated with a higher likelihood of HIV infection . Another plausible explanation is that occupations more likely to experience needlestick exposures are held more frequently by women , potentially resulting in a higher number of women-related tests for occupational exposure. Finally, an ED study on victims of sexual offense found that 84% of victims were women .
The reason for the observed temporal trends in HIV testing is not clear. Explanations are likely multifactorial and include the dynamics of the US HIV/AIDS epidemic, changing CDC recommendations for appropriate testing sites, evolution of HIV testing methods, changes in federal and state policy, emergency medicine community's perception of public health interventions, individual ED policies, and resource and operational issues specific to EDs (e.g. ED overcrowding). It is interesting to note that the highest two testing periods occurred immediately after the release of the 1993 and 2001 recommendations. Although it is possible that the recommendations served as a stimulus for providers to order more tests, there are no direct data to support this. A possible explanation for the significant drop in rates during the years 1997–99 is that the introduction of highly active antiretroviral therapy and the associated reduction in AIDS-related morbidity and mortality led the emergency medicine community to be less vigilant with regard to HIV testing recommendations. It is also plausible that providers began to encounter fewer patients with HIV/AIDS-related illness and were accordingly less attentive to the possibility of unrecognized HIV in the populations they served. Another possible explanation for the observed trend is that increased availability of traditional testing venues and community outreach programs led to a relative decline in ED-based testing. Finally, the Food and Drug Administration approved simple, easy to use, point-of-care rapid HIV tests in 2002 . Observed increased rates of testing seen in EDs might accordingly be attributed to increased acceptance of a test that could be more easily integrated into routine ED practice . Several EDs have demonstrated that implementing an ED-based rapid HIV testing program is feasible and able to identify significant number of individuals with previously unrecognized infection [19,21,22,51]. One study by Lubelchek et al.  showed that ED-based rapid testing for admitted patients could shorten the hospital stay by 5.4 days and the interval between discharge and initial outpatient follow-up for HIV infection by 28 days as compared with conventional in-patient testing.
The extremely high ORs of receiving an HIV test for ED visits for puncture wound/needlestick, exposure to bodily fluids of another person, and rape are logical and expected. According to CDC recommendations, each of these represents an indicator condition that should prompt clinicians to test. Of note, however, the overall rate of testing in EDs for each of these groups was relatively low. These data are consistent with those reported elsewhere [23,25].
The two other conditions that were independently associated with an increased likelihood of obtaining an HIV test in EDs were fever and mental/psychological disorders. Fever is a common symptom presenting in those with both acute and chronic infections. It is possible that the presence of fever in the appropriate clinical context (e.g. recurrent bacterial pneumonia, suspected pneumocystis pneumonia, tuberculosis, esophagitis, and meningitis) triggered ED providers to order an HIV test as part of their diagnostic evaluation. Additionally, fever may have prompted physicians to consider acute HIV infection, as previous reports have shown that more than 70% of those with acute HIV infection manifest fevers . Of note, however, the current, commonly used HIV enzyme immunoassay [54,55] is incapable of diagnosing acute HIV infection. As a result, ED-based screening for acute seroconversion with HIV RNA viral load has been advocated by some and is already demonstrated to be a successful strategy for early identification of the disease .
Mental/psychological disorders have been correlated with behaviors (especially substance abuse and high-risk sexual practices) known to be associated with increased risk of HIV transmission  and a higher prevalence of HIV infection [58–60]. In addition, symptoms of mental/psychological disorders such as confusion, forgetfulness, or behavioral changes frequently occur as a result of HIV-related opportunistic infections, tumors in the central nervous system, or direct neuropathogenic effects of HIV [59,61]. Our findings suggest that ED providers may see mental/psychological disorders as a marker of higher risk for unrecognized HIV and order HIV tests for these patients at a higher rate.
Some infectious diseases, for instance, pneumonia, tuberculosis, and STDs, frequently coexist with HIV. However, in the present study, we did not find that symptoms/signs of these infectious diseases, such as cough, chest pain, abdominal pain, or urogenital discharge, triggered a higher HIV testing rate. As noted above, our findings are consistent with the conclusion of ED providers' survey studies, that is, low perceived HIV testing rate in patients with STDs [23–26]. It has not yet been reported in the literature whether having an ED diagnosis of pneumonia or tuberculosis signals a perceived need or actual increase in testing for HIV. As discussed above, only 36% of ED patients with suspected AIDS-defining illness were prompted to test for HIV .
In our study, a substantial proportion, 6%, of those tested were diagnosed with HIV infection. Although this database does not allow us to definitively state whether these represent newly recognized infections, most likely do, as an HIV test was ordered during that ED visit. This rate is in keeping with findings of other investigators [10–12,15,17–19] and likely indicates that however low the rates of testing actually are, ED-based programs are effective at detecting new infections.
The present study has several limitations. First, the deidentified NHAMCS database does not reveal underlying endemic rates of HIV infection. Thus, the denominator data regarding ED visits include sources that may not meet the CDC threshold for offering testing in ED settings, thus, artificially decreasing the testing rate. We do not know to what extent such sites affected the data if at all. Second, NHAMCS does not collect HIV risk profiles for patient visits. Consequently, we cannot say with certainty whether testing was driven by clinical presentation, high-risk behavior, or both. Third, we are unable to ascertain whether ED providers referred patients to other testing venues for HIV testing. Fourth, it is possible that routine or targeted testing was offered, but some patients were not able to provide consent for testing or declined to be tested.
In conclusion, we used NHAMCS data to provide the first national baseline estimates of rates of HIV testing in US EDs for over a recent 13-year period, preceding the release of the most recent CDC expanded HIV testing recommendations. Only a small minority of visits was associated with HIV testing even after 2001, when the CDC explicitly recommended that EDs meeting certain threshold criteria should offer routine HIV CTR. However infrequent, current testing appears to be driven by the patient's clinical presentation in EDs. Given the potential role of EDs in the national strategy for HIV reduction, further research should be directed at discovering and alleviating reasons for low rates of HIV testing in ED settings.
The authors would like to thank Mr. Ting-Yu Shih for his assistance with data summary. R.E.R. has been supported in part by a grant from Maryland Department of Health and Mental Hygiene.
Author contributions: Y.-H.H. has full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Y.-H.H. Analysis and interpretation of data: Y.-H.H., R.E.R., D.E.N.-T., G.D.K. Drafting of the manuscript: Y.-H.H. Critical revision of the manuscript for important intellectual content: Y.-H.H., R.E.R., D.E.N.-T., G.D.K. Statistical analysis: Y.-H.H. Study supervision: R.E.R.
1. Glynn M, Lee L, McKenna M. The status of national HIV case surveillance, United States 2006. Public Health Rep 2007; 122:63–71.
3. Institute of Medicine. No time to lose: getting more from HIV prevention
. Washington DC: National Academy Press; 2001.
4. Marks G, Crepaz N, Senterfitt J, Janssen R. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: implications for HIV prevention programs. J Acquir Immune Defic Syndr 2005; 39:446–453.
5. Centers for Disease Control and Prevention. Recommendations for HIV testing services for inpatients and outpatients in acute care hospital settings
. MMWR Recomm Rep
6. Centers for Disease Control and Prevention. Revised recommendations for HIV counseling and testing
. MMWR Recomm Rep
7. Branson B, Handsfield H, Lampe M, Janssen RS, Taylor AW, Lyss SB, et al
. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 2006; 55:1–17.
8. Centers for Disease Control and Prevention (CDC). Missed opportunities for earlier diagnosis of HIV infection: South Carolina, 1997–2005.MMWR Morb Mortal Wkly Rep
9. Liddicoat R, Horton N, Urban R, Maier E, Christiansen D, Samet J. Assessing missed opportunities for HIV testing in medical settings. J Gen Intern Med 2004; 19:349–356.
10. Kelen G, Shahan J, Quinn T. Emergency department-based HIV screening and counseling: experience with rapid and standard serologic testing. Ann Emerg Med 1999; 33:147–155.
11. Schoenbaum E, Webber M. The underrecognition of HIV infection in women in an inner-city emergency room. Am J Public Health 1993; 83:363–368.
12. Kelen G, Hexter D, Hansen K, Humes R, Vigilance PN, Baskerville M, Quinn TC. Feasibility of an emergency department-based, risk-targeted voluntary HIV screening program. Ann Emerg Med 1996; 27:687–692.
13. Kelen G, Hexter D, Hansen K, Tang N, Pretorius S, Quinn T. Trends in human immunodeficiency virus (HIV) infection among a patient population of an inner-city emergency department: implications for emergency department-based screening programs for HIV infection. Clin Infect Dis 1995; 21:867–875.
14. Lindsay M, Grant J, Peterson H, Risby J, Williams H, Klein L. Human immunodeficiency virus infection among patients in a gynecology emergency department. Obstet Gynecol 1993; 81:1012–1015.
15. Goggin M, Davidson A, Cantril S, O'Keefe L, Douglas J. The extent of undiagnosed HIV infection among emergency department patients: results of a blinded seroprevalence survey and a pilot HIV testing program. J Emerg Med 2000; 19:13–19.
16. Alpert P, Shuter J, DeShaw M, Webber M, Klein R. Factors associated with unrecognized HIV-1 infection in an inner-city emergency department. Ann Emerg Med 1996; 28:159–164.
17. Brown J, Shesser R, Simon G, Bahn M, Czarnogorski M, Kuo I, et al
. Routine HIV screening in the emergency department using the new US Centers for Disease Control and Prevention Guidelines: results from a high-prevalence area. J Acquir Immune Defic Syndr 2007; 46:395–401.
18. White D, Cheung P, Scribner A, Alter H. A comparison of the feasibility and yield of routine rapid HIV testing in an urban emergency department and urgent care center. Acad Emerg Med 2007; 14:S14.
19. Lyss S, Branson B, Kroc K, Couture E, Newman D, Weinstein R. Detecting unsuspected HIV infection with a rapid whole-blood HIV test in an urban emergency department. J Acquir Immune Defic Syndr 2007; 44:435–442.
20. Lyons M, Lindsell C, Ledyard H, Frame P, Trott A. Emergency department HIV testing and counseling: an ongoing experience in a low-prevalence area. Ann Emerg Med 2005; 46:22–28.
21. Mehta S, Hall J, Lyss S, Skolnik P, Pealer L, Kharasch S. Adult and pediatric emergency department sexually transmitted disease and HIV screening: programmatic overview and outcomes. Acad Emerg Med 2007; 14:250–258.
22. Silva A, Glick N, Lyss S, Hutchinson AB, Gift TL, Pealer LN, et al
. Implementing an HIV and sexually transmitted disease screening program in an emergency department. Ann Emerg Med 2007; 49:564–572.
23. Ehrenkranz P, Ahn C, Metlay J, Camargo CJ, Holmes W, Rothman R. Availability of rapid human immunodeficiency virus testing in academic emergency departments. Acad Emerg Med 2008; 15:144–150.
24. Gift T, Hogben M. Emergency department sexually transmitted disease and human immunodeficiency virus screening: findings from a national survey. Acad Emerg Med 2006; 13:993–996.
25. Wilson S, Mitchell C, Bradbury D, Chavez J. Testing for HIV: current practices in the academic ED. Am J Emerg Med 1999; 17:354–356.
26. Fincher-Mergi M, Cartone K, Mischler J, Pasieka P, Lerner E, Billittier A. Assessment of emergency department healthcare professionals' behaviors regarding HIV testing and referral for patients with STDs. AIDS Patient Care STDS 2002; 16:549–553.
27. Stussman B. National Hospital Ambulatory Medical Care Survey: 1995 emergency department summary. Adv Data 1997; 285:1–19.
28. McCaig L, Stussman B. National Hospital Ambulatory Medical Care Survey: 1996 emergency department summary. Adv Data 1997; 293:1–20.
29. Nourjah P. National Hospital Ambulatory Medical Care Survey: 1997 emergency department summary. Adv Data 1999; 304:1–24.
30. McCaig L. National Hospital Ambulatory Medical Care Survey: 1998 emergency department summary. Adv Data 2000; 313:1–23.
31. McCaig L, Burt C. National Hospital Ambulatory Medical Care Survey: 1999 emergency department summary. Adv Data 2001; 320:1–34.
32. McCaig L, Burt C. National Hospital Ambulatory Medical Care Survey: 2001 emergency department summary. Adv Data 2003; 335:1–29.
33. McCaig L, Burt C. National Hospital Ambulatory Medical Care Survey: 2002 emergency department summary. Adv Data 2004; 340:1–34.
35. Centers for Disease Control and Prevention. NHAMCS micro-data file documentation
. Atlanta, Georgia: Centers for Disease Control and Prevention; 2003. pp. 1–184.
36. Schneider D, Appleton L, McLemore T. A reason for visit classification for ambulatory care
. Vital Health Stat
2 1979; 78:1–63.
39. Kelen G, Fritz S, Qaqish B, Brookmeyer R, Baker JL, Kline RL, et al
. Unrecognized human immunodeficiency virus infection in emergency department patients. N Engl J Med 1988; 318:1645–1650.
40. Rothman R. Current Centers for Disease Control and Prevention guidelines for HIV counseling, testing, and referral: critical role of and a call to action for emergency physicians. Ann Emerg Med 2004; 44:31–42.
41. Rothman R, Ketlogetswe K, Dolan T, Wyer P, Kelen G. Preventive care in the emergency department: should emergency departments conduct routine HIV screening? A systematic review. Acad Emerg Med 2003; 10:278–285.
42. Burke R, Sepkowitz K, Bernstein K, Karpati AM, Myers JE, Tsoi BW, Begier EM. Why don't physicians test for HIV? A review of the US literature. AIDS 2007; 21:1617–1624.
43. Brown J, Shesser R, Simon G. Establishing an ED HIV Screening Program: lessons from the front lines. Acad Emerg Med 2007; 14:658–661.
44. Kelen G. Public health initiatives in the emergency department: not so good for the public health? Acad Emerg Med 2008; 15:194–197.
46. Fenton K. Changing epidemiology of HIV/AIDS in the United States: implications for enhancing and promoting HIV testing strategies. Clin Infect Dis 2007; 45:S213–S220.
47. Slaven E, Lopez F, Weintraub S, Mena J, Mallon W. The AIDS patient with abdominal pain: a new challenge for the emergency physician. Emerg Med Clin North Am 2003; 21:987–1015.
48. US Department of Labor. Employment and Earnings
. Washington, DC: Bureau of Labor Statistics; 2001. http://www.bls.gov/cps/home.htm
. [Accessed 7 April 2008].
49. Straight J, Heaton P. Emergency department care for victims of sexual offense. Am J Health Syst Pharm 2007; 64:1845–1850.
50. Branson B. State of the art for diagnosis of HIV infection. Clin Infect Dis 2007; 45:S221–S225.
51. Centers for Disease Control and Prevention (CDC). Rapid HIV testing in emergency departments: three U.S. sites, January 2005–March 2006.MMWR Morb Mortal Wkly Rep
52. Lubelchek R, Kroc K, Hota B, Sharief R, Muppudi U, Pulvirenti J, Weinstein RA. The role of rapid vs conventional human immunodeficiency virus testing for inpatients: effects on quality of care. Arch Intern Med 2005; 165:1956–1960.
53. Zetola N, Pilcher C. Diagnosis and management of acute HIV infection. Infect Dis Clin North Am 2007; 21:19–48.
54. Fiebig E, Wright D, Rawal B, Garrett PE, Schumacher RT, Peddada L, et al
. Dynamics of HIV viremia and antibody seroconversion in plasma donors: implications for diagnosis and staging of primary HIV infection. AIDS 2003; 17:1871–1879.
56. Pilcher C, Fiscus S, Nguyen T, Foust E, Wolf L, Williams D, et al
. Detection of acute infections during HIV testing in North Carolina. N Engl J Med 2005; 352:1873–1883.
57. Desai M, Rosenheck R. HIV testing and receipt of test results among homeless persons with serious mental illness. Am J Psychiatry 2004; 161:2287–2294.
58. Rosenberg S, Goodman L, Osher F, Swartz MS, Essock SM, Butterfield MI, et al
. Prevalence of HIV, hepatitis B, and hepatitis C in people with severe mental illness. Am J Public Health 2001; 91:31–37.
59. Baillargeon J, Ducate S, Pulvino J, Bradshaw P, Murray O, Olvera R. The association of psychiatric disorders and HIV infection in the correctional setting. Ann Epidemiol 2003; 13:606–612.
60. Beyer J, Taylor L, Gersing K, Krishnan K. Prevalence of HIV infection in a general psychiatric outpatient population. Psychosomatics 2007; 48:31–37.
61. Adams M, Ferraro F. Acquired immunodeficiency syndrome dementia complex. J Clin Psychol 1997; 53:767–778.