Falling through the cracks? Missed opportunities for earlier HIV diagnosis in a New York City Hospital

Abstract

Summary

Newly diagnosed HIV-positive patients have frequent health care encounters prior to diagnosis representing missed opportunities for diagnosis. This study determines the proportion of patients with new HIV diagnoses with encounters in the 3 years prior to diagnosis. We describe the characteristics of newly diagnosed patients and of “late testers” (CD4 <200 cells/mm³ at the time of diagnosis). We identified all newly diagnosed with HIV in emergency department, inpatient, and outpatient settings between May 1, 2006, and December 31, 2009. Data abstractors searched hospital records to identify all emergency department, inpatient, and outpatient visits for the 3 years prior to diagnosis. In all, 23,271 HIV tests were performed and 253 persons were newly diagnosed (1.1%); 152 new positives (60.1%) made at least one prior visit. Of patients with CD4 counts available, 104/175 (59.4%) had CD4 <200 cells/mm³. Patients with at least one prior visit had a median of three. There was no difference in numbers of visits between late testers and non-late testers, although late testers were more likely to have ED visits. Most newly diagnosed HIV-positive patients had multiple encounters prior to diagnosis. Many of these patients presented with CD4 counts below 200 cells/mm³, indicating true missed opportunities for earlier diagnosis.

Keywords

Missed opportunities HIV testing HIV/AIDS diagnosis screening late presenters late testers

Introduction

Over one million people in the United States (US) are estimated to be HIV positive, with 56,000 estimated new infections occurring each year.¹ Moreover, Centers for Disease Control and Prevention (CDC) data estimate that approximately 18.3% of HIV-positive patients do not know they are infected.² Estimates are that persons unaware of their status account for the transmission of 54–70% of new infections.^3–5 Substantial work has shown that knowledge of HIV infection leads people to adopt safer behaviours,^4,6–8 suggesting that HIV testing may be an effective intervention in preventing spread of the disease.^9,10 Early detection also allows access to earlier treatment and care,^7,11–13 including highly-active anti-retroviral therapy (HAART) which decreases both morbidity and early mortality.^14,15

In 2006, the CDC published guidelines recommending opt-out HIV screening in all individuals ages 13–64 presenting for care in any health care setting.¹⁰ This recommendation was partially based on data from South Carolina collected over a 9-year period that suggest many people with undiagnosed HIV infection present to health care facilities multiple times without being diagnosed, as they present for complaints that would not prompt HIV testing under a risk-based testing strategy.¹⁶ Furthermore, the US Preventive Services Task Force recently recommended expansion of routine testing to all individuals between 15 and 65.¹⁷ Nonetheless, there is continuing debate about the role and effectiveness of non-targeted HIV testing in health care settings, including emergency departments (EDs).^18–21

Our objectives were twofold: to determine the proportion of patients with a new diagnosis of HIV at an urban tertiary hospital in New York City who presented to our hospital system in the 3 years prior to diagnosis, and to describe the characteristics of newly diagnosed patients and of “late testers” (patients with CD4 < 200 cells/mm³ within 100 days of HIV diagnosis) in whom prior health care encounters represent clear examples of missed opportunities for earlier diagnosis.

Materials and methods

To achieve our objectives, we conducted a chart review of patient records at St. Luke's-Roosevelt Hospital Center (SLRHC) in Manhattan, New York City, of all patients who tested newly positive for HIV infection between May 1, 2006, and December 31, 2009. Electronic records were searched to characterize newly positive patients, record their CD4 count within 100 days of their positive test, and identify their ED, inpatient, and outpatient visits for the 3 years prior to their diagnosis. Only visits to SLRHC were included in the analysis.

SLRHC is a 1076-bed university-affiliated hospital serving a diverse, urban population comprising two campuses on the west side of Manhattan approximately three miles apart. In 2009, the SLRHC ED saw approximately 187,000 visits. Rapid HIV testing has been available since 2006 in both EDs and at affiliated inpatient and outpatient facilities including inpatient wards, labour and delivery, outpatient medical clinics, obstetrics and gynaecology clinics, community primary care clinics, and outpatient HIV clinics. The methods by which testing is initiated varies by venue (e.g. non-targeted, diagnostic, patient request). All positive patients in the outpatient clinics or ED are referred for follow-up at the outpatient HIV clinics on the next business day or immediately following their inpatient stay. Using automated reports of positive HIV tests performed in the hospital, a trained social worker also contacts the patient by phone the day after the positive test in the ED or clinics or meets them in the hospital during their inpatient stay to further encourage successful linkage to care.

Included in this study were individuals aged 16 and older who tested HIV positive for the first time at one of the above sites between May 1, 2006, and December 31, 2009. By New York State (NYS) law during the study period, all patients signed a written informed consent for HIV testing prior to test administration.

Research assistants were trained in data extraction for the purpose of this study using a standardized data collection tool formatted with all terms defined before abstraction. Assistants were also trained in confidentiality guidelines and signed confidentiality agreements. Data were extracted and tabulated in a secured location. Assistants were blinded from the aim and objectives of this study. The principal investigators periodically monitored the data collection and re-abstracted chart data for crosschecking.

Data were stored in a password-protected file. Each patient was assigned a specific identification number for this study, and medical record numbers were eliminated from the data set after collection was complete. The primary investigators (who conducted the data analysis) did not have access to any personal identifiers of the study participants, with the exception of a small proportion of data for the purposes of quality control. Confidentiality was maintained at all times.

Data for mandatory reporting to the New York State Department of Health (NYS DOH) have been maintained since the initiation of rapid HIV testing at SLRHC in 2006. These data include age, sex, race/ethnicity, and date of positive test and provided the initial data for collection. Additional data were accessed through the hospital electronic medical record systems including insurance status, CD4 count, prior SLRHC visits, and specific sites of visits. Medical records were reviewed for patient visit history from May 2006 (when testing began) until the time of new HIV diagnosis for a maximum of 3 years. Visits were counted and considered “missed opportunities” if the individual's visit was in 2006 or later, during which rapid HIV testing was available throughout SLRHC. Those patients with a CD4 count of less than 200 cells/mm³ within 100 days of their diagnostic visit were defined as “late-testers.”

Data exploration was conducted using Microsoft Excel 2007 (Microsoft Corporation, Redmond, WA) and counts, percentages, and medians were calculated. The z-test was used to compare proportions and the Wilcoxon-Mann-Whitney test to compare medians. p Values were calculated at the alpha = 0.05 level of significance. The Institutional Review Board of SLRHC approved the study.

Results

Over the 4-year study period, 23,271 individuals were tested for HIV (Table 1); 253 were found to be HIV positive, for a percentage of 1.1%. Of these, 175 had CD4 counts recorded within 100 days, and 104 (59.4%) were late testers, i.e. had a CD4 count less than 200 cells/mm³; 127 (72.6%) had a CD4 count less than 350 cells/mm³.

Table 1.

Testing, HIV positivity, and CD4 counts by setting, St. Luke's-Roosevelt Hospital Centre, May 1, 2006, and December 31, 2009.

	ED	Inpatient	Outpatient	Total	p Value
Number of HIV tests performed	8448	4863	9960	23,271	–
Number of new positives (%)	132 (1.6%)	79 (1.6%)	42 (0.4%)	253 (1.1%)	ED vs. inpatient: 1.00ED vs. outpatient: <0.01Inpatient vs. outpatient: <0.01
Number of new positives with CD4 count on record within 100 days of testing	89 (67.4%)	63 (79.7%)	23 (54.8%)	175 (69.2%)	ED vs. inpatient: 0.05ED vs. outpatient: 0.14Inpatient vs. outpatient: <0.01
Median CD4 count (cells/mm³) of new positives (interquartile range)	89 (20–346) (n = 89)	83 (28–314) (n = 63)	370 (155–470) (n = 23)	122 (26–365) (n = 175)	ED vs. inpatient: 0.84ED vs. outpatient: <0.01Inpatient vs. outpatient: <0.01
Number of late testers, with CD4 count <200 cells/mm³ (%)	54 (60.7%) (n = 89)	43 (68.3%) (n = 63)	7 (30.4%) (n = 23)	104 (59.4%) (n = 175)	ED vs. inpatient: 0.33ED vs. outpatient: <0.01Inpatient vs. outpatient: <0.01
Number of testers with CD4 count <350 cells/mm³ (%)	67 (75.3%) (n = 89)	49 (77.8%) (n = 63)	11 (47.8%) (n = 23)	127 (72.6%) (n = 175)	ED vs. inpatient: 0.72ED vs. outpatient: 0.01Inpatient vs. outpatient: 0.01
Number of testers with CD4 count <500 cells/mm³ (%)	77 (86.5%) (n = 89)	58 (92.1%) (n = 63)	19 (82.6%) (n = 23)	154 (88.0%) (n = 175)	ED vs. inpatient: 0.28ED vs. outpatient: 0.63Inpatient vs. outpatient: 0.20

The median age of those testing newly positive was 40 years, though late testers tended to be older than non-late testers (Table 2); 175(69.2%) of the newly positive were men, greater than half identified as black, and over a quarter identified as Hispanic. A majority of positive testers were either uninsured or on Medicaid, a US government health insurance program regulated on a state level for low-income individuals and families.

Table 2.

Characteristics, prior visits, and location of visits of persons testing newly HIV positive, St. Luke's-Roosevelt Hospital Center, May 1, 2006, and December 31, 2009.

	All new positives (n = 253)	Late testers (CD4 <200; n = 104)	Non-late testers (CD4 > = 200; n = 71)	p Value
Median age (interquartile range)	40 years (28–47)	43 years (33–50)	39 years (23–47)	0.02
Sex
Men	69.2% (n = 175)	75.0% (n = 78)	62.8% (n = 44)	0.09
Women	30.8% (n = 78)	25.0% (n = 26)	37.1% (n = 26)	0.09
Race/ethnicity
White	9.5% (n = 24)	11.5% (n = 12)	8.5% (n = 6)	0.51
Black	58.9% (n = 149)	54.8% (n = 57)	61.9% (n = 44)	0.35
Hispanic	25.7% (n = 65)	26.9% (n = 28)	23.9% (n = 17)	0.66
Other/Unknown	5.9% (n = 15)	6.7% (n = 7)	5.6% (n = 4)	0.77
Primary insurance
Commercial insurance	22.5% (n = 57)	25.0% (n = 26)	16.9% (n = 12)	0.20
Medicare^a	6.3% (n = 16)	11.5% (n = 12)	4.2% (n = 3)	0.09
Medicaid^b	41.1% (n = 104)	39.4% (n = 41)	47.9% (n = 34)	0.27
Uninsured	27.7% (n = 70)	21.2% (n = 22)	29.6% (n = 21)	0.20
Unknown	2.4% (n = 6)	2.9% (n = 3)	1.4% (n = 1)	0.52
Prior visits of persons testing newly positive
Percent who made at least one visit for medical care in the 3 years prior to diagnosis	60.1% (n = 152/253)	57.7% (n = 60/104)	67.6% (n = 48/71)	0.19
Total prior visits	1025	438	386	–
Median number (interquartile range) of prior visits in those with at least one	3.0 (1.0–7.0)	4.0 (2.0–9.3)	2.5 (1.0–5.3)	0.16
Median (interquartile range) duration between first visit within a 3 year period and date of diagnosis	251 days (32–624)	270 days (53–528)	239 days (19–659)	0.93
Location of visits
Emergency Department	40.6% (n = 416/1025)	43.2% (n = 189/438)	28.5% (n = 110/386)	<0.01
Outpatient facilities	49.9% (n = 511/1025)	47.5% (n = 208/438)	62.4% (n = 241/386)	<0.01
Inpatient admissions	9.6% (n = 98/1025)	9.4% (n = 41/438)	9.1% (n = 35/386)	0.88

^aMedicare is a United States government health insurance program for persons 65 years old and older, younger persons with disabilities, persons with end-stage renal disease, and persons with Amyotrophic Lateral Sclerosis.

^bMedicaid a United States government health insurance programme regulated on a state-level for low-income individuals and families.

In all, 152, or more than 60%, of individuals newly diagnosed with HIV had made at least one visit to the SLRHC for medical care in the time during which HIV testing was available (measured up to three years prior to testing positive for HIV) (Table 2). The median number of prior visits was three; the median duration between first visit and date of diagnosis was 251 days. Late testers did not make significantly more visits prior to diagnosis than non-late testers. Late testers were diagnosed more often in EDs than non-late testers, and correspondingly less in outpatient facilities than non-late testers (Table 2).

Discussion

The percentage of HIV (1.1%) among all 23,271 individuals who consented for rapid HIV testing during this 4-year study period is more than double that of the national 2010 prevalence, 0.37%,²² but approximates the 2010 prevalence of HIV within New York City, 1.36%.^23,24

In our study, nearly 60% of those with CD4 counts recorded within 100 days of testing had a CD4 count less than 200 cells/mm³. This proportion is greater than the 38.3% of persons nationally that are newly diagnosed with HIV and AIDS within 1 year.¹³ This is a higher proportion than shown in prior studies as well. During 2001–2005, of all newly positive HIV cases in South Carolina, AIDS was diagnosed in 41% of individuals within 1 year of their initial HIV diagnosis, including 16.5% in whom AIDS was diagnosed within 30 days of their initial HIV diagnosis.² Among HIV screening in a Chicago ED over a 15-month period 2003–2004, 45.2% of those testing newly positive had AIDS. Finally, 43% of those testing newly positive in a Californian HMO had CD4 counts consistent with AIDS.¹⁹ Interestingly, our proportion of late testers is also substantially greater than general New York City statistics, which suggest than 21% of those newly diagnosed with HIV in 2010 were concurrently diagnosed with AIDS.²⁵ Several reasons may explain the higher proportion of late testers that we found. First, our urban ED population may suffer from a disproportionate lack of access to routine health care services limiting the generalizability of this work to other centers. Individuals may wait until a dire illness to seek medical care, and hence do so in acute care settings such as our EDs. Additionally, patients may have encountered health care providers at times when robust routine testing efforts were not implemented. Our proportions are likely influenced by the fact that testing was not routine in all settings, despite CDC's 2006 guidelines recommending opt-out screening for all individuals aged 13–64, and provider suspicion may have targeted testing for certain patients based on clinical suspicion (e.g. pneumonia suspicious for Pneumocystis). All of our patient encounters occurred prior to the 2010 NYS legislation mandating the offer of an HIV test to any person aged 13–64 presenting for health care services. It would be of interest to repeat this study evaluating a 3-year period during which the offer of an HIV test was required by law to determine whether the proportion of late-testers would differ.

This US phenomenon of missed opportunities in HIV testing reflects a common global theme: works have described missed opportunities in Ethiopia, France, Ireland, Kenya, and Scotland in recent years.^26–31 Deblonde reports an estimated 30% of people living with HIV in the European Union and neighbouring countries are unaware of their infection and suggests missed opportunities for diagnosis due to a variety of factors including patient perceptions, legal or financial factors, and attitudes and practices of health care providers.³² HIV testing guidelines have been introduced in Scotland and the United Kingdom; similarly to the US, both countries have had challenges in applying these guidelines in practice.^31,33,34

The findings of one US study conducted suggest that people diagnosed with HIV during ED visits tend to have lower CD4 counts than those diagnosed in other health care settings,³⁵ and the CDC suggests that EDs may have a role in earlier HIV diagnosis. Our findings support this recommendation, as late testers are more likely than non-late testers to be diagnosed in the ED. Further, we found persons testing in EDs and inpatient settings were more likely to test positive that persons testing as inpatients, and ED and inpatient testers were more likely to have lower CD4 counts within 100 days of testing than outpatient testers. Of prior health care visits made by late testers in South Carolina, 78.9% were to EDs.¹⁶ Of all newly diagnosed positives, over 40% were diagnosed in our EDs, a number consistent with data from the Massachusetts cohort in which 40% of missed opportunities were in ED or urgent care settings.⁷ We present counts of late testers, those with CD4 counts less than 200 cells/mm³ and by definition with AIDS, to illustrate the potential impact that routine opt-out HIV testing in EDs can make on health care outcomes. An even greater proportion of patients would have benefitted from earlier identification and access to care because antiretroviral therapy (ART) is now strongly recommended for HIV-positive individuals with CD4 counts </ = 500 with CD4 counts cells/mm³ and additionally recommended for individuals with CD4 counts even higher and in specific situations of high transmission risk.^15,36

Inpatients testing positive were more likely to have a CD4 count recorded within 100 days of testing than ED patients or outpatients testing positive. This was likely because follow-up testing is essentially guaranteed for inpatients during the admission or immediately after. Out of those testing positive, CD4 counts were not recorded on 58 newly positive individuals, and 17 individuals had CD4 counts recorded more than 100 days after diagnosis. This may suggest either poor follow-up and unsuccessful linkage to care or perhaps a separate database in which this information may have been recorded, such as a facility outside of our network. However, the presence of a CD4 count may not be a fully accurate surrogate for follow-up since it could have been sent during their inpatient stay or with a confirmatory test and may not indicate full engagement in care. We are unable to deduce whether the subset of patients with CD4 counts recorded in our system is representative of all new positives as a whole.

Overall, late testers were older than non-late testers. Nearly 70% of individuals testing newly positive in our study were men; this is reflective of national incidence statistics showing 74–76% of those testing newly positive throughout the US are men.³⁷ Approximately 26% of our incident cases identified as being Hispanic; a number consistent with the 18–20% of incident cases being among Hispanics nationally. In contrast, 58.9% of our incident cases were among those identifying as black, which significantly exceeds the national proportion of 42–46%.³⁷ However, this does correlate more closely with the rates of citywide infections in NYC. In 2006, 50.7% of new HIV cases in NYC were black compared to 48% in 2009.²⁴ During 2009, 32% of patients evaluated in the SLRHC ED were black, indicating that the percentage of new black positives was not based solely on the patient demographics. The entire demographic profile of individuals tested is not available for the study period and therefore we cannot speculate on whether our finding of more men than women positives was a result of more men having been tested. However, our hospital system serves an area of NYC with high-risk male patients, i.e. minorities in the Harlem community as well as men who have sex with men (MSM) patients from the Chelsea and Hell's Kitchen neighbourhoods where the prevalence of HIV is relatively high.

Several studies have been published in recent years on missed opportunities for early diagnosis of HIV infection in a variety of settings.^{7,14,16,38–40} In our study, over 60% of those testing newly positive for HIV made at least one visit for medical care in the 3 years prior to their diagnosis. These visits likely represent missed opportunities to diagnose HIV infection at an earlier stage. This finding represents a proportion that is higher than in three prior studies reporting missed opportunities: 27.5%,³⁵ 30.5%,³⁸ and 48%.⁴¹ We cannot fully explain this difference other than to suggest that our significantly underinsured population tends to use the hospital-based system including the ED for more routine, non-urgent issues.

In our study, individuals in whom opportunities for diagnosis were missed had a median number of three prior visits over a median duration of 251 days. In comparison, among a Southern Californian cohort of newly positive individuals who had made prior health care visits, the median delay in diagnosis was 112 days.¹⁴ A 10-year review of newly positive individuals in Massachusetts revealed a median number of five visits per patient prior to diagnosis. All of these studies will be dependent on patient populations, access to care, and the type of setting in which they have medical encounters. It is possible that rates of missed opportunities may decrease in NYS with the new testing legislation. However, this opt-out model of testing still requires the patient to accept a test, which will always be a limiting factor in detection.

Limitations

Demographic information such as age, sex, race/ethnicity, and primary insurance were not collected on all persons consenting for an HIV test at our centres; hence, we are unable to compare persons testing negative with those testing newly positive on these characteristics. Further, we were unable to include in this work health facility visits an individual may have had prior to testing positive outside of our hospital network. Finally, we only included visits within 3 years prior to diagnosis and visits that occurred in 2006 or later. Hence, our calculation of prior visits can only be an underestimation of opportunities missed.

The testing model in each of the clinical settings was not identical and patient-declined testing opportunities may have occurred. We were not able to capture these data. This work took place prior to the NYS law mandating the offer of an HIV test to all patients aged 13–64 presenting for health care services.

Conclusion

Most newly diagnosed HIV-positive patients in our study had multiple health care encounters prior to their diagnosis. A large proportion of these presented with AIDS, indicating that even in developed counties multiple opportunities to identify HIV infection, engage patients in care, and potentially reduce spread of the disease were missed. Since earlier ART is now recommended for HIV-positive individuals an even greater proportion of patients would have benefitted from earlier identification and access to care. Late testers were significantly more likely to have had a previous visit to the ED than non-late testers. This work supports increased efforts to implement more expanded HIV testing in health care settings including the ED as part of a comprehensive public health strategy to decrease morbidity and mortality, and reduce spread of the virus.

Footnotes

Acknowledgements

The test kits used in this study were funded by a grant from the New York City Department of Health.

Conflict of interest

The authors declare no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Hall

Song

Rhodes

. Estimation of HIV incidence in the United States. JAMA 2008; 300: 520–529.

Batey

Hogan

Cantor

. Short communication routine HIV testing in the emergency department: assessment of patient perceptions. AIDS Res Hum Retroviruses 2012; 28: 352–356.

Freeman

Sattin

Miller

. Acceptance of rapid HIV screening in a southeastern emergency department. Acad Emerg Med 2009; 16: 1156–1164.

Marks

Crepaz

Janssen

. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS Lond Engl 2006; 20: 1447–1450.

Centers for Disease Control and Prevention (CDC). HIV prevalence estimates--United States, 2006. MMWR Morb Mortal Wkly Rep 2008; 57: 1073–1076.

Berrios

Hearst

Coates

. HIV antibody testing among those at risk for infection. The National AIDS Behavioral Surveys. JAMA 1993; 270: 1576–1580.

Liddicoat

Horton

Urban

. Assessing missed opportunities for HIV testing in medical settings. J Gen Intern Med 2004; 19: 349–356.

Higgins

Galavotti

O'Reilly

. Evidence for the effects of HIV antibody counseling and testing on risk behaviors. JAMA 1991; 266: 2419–2429.

Haukoos

Lyons

. Idealized models or incremental program evaluation: translating emergency department HIV testing into practice. Acad Emerg Med 2009; 16: 1044–1048.

10.

Branson

Handsfield

Lampe

. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 2006; 55: 1–17.

11.

HIV testing and screening in the emergency department. Ann Emerg Med 2007; 50: 209–209.

12.

Late versus early testing of HIV--16 Sites, United States, 2000–2003. MMWR Morb Mortal Wkly Rep 2003; 52: 581–586.

13.

Late HIV testing - 34 states, 1996–2005. MMWR Morb Mortal Wkly Rep 2009; 58: 661–665.

14.

Kuo

Haukoos

Witt

. Recognition of undiagnosed HIV infection: an evaluation of missed opportunities in a predominantly urban minority population. AIDS Patient Care STDs 2005; 19: 239–246.

15.

Kitahata

Gange

Abraham

. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med 2009; 360: 1815–1826.

16.

Missed opportunities for earlier diagnosis of HIV infection--South Carolina, 1997–2005. MMWR Morb Mortal Wkly Rep 2006; 55: 1269–1272.

17.

US Preventive Services Task Force. Screening for Human Immunodeficiency Virushttp://www.uspreventiveservicestaskforce.org/uspstf/uspshivi.htm (accessed 11 Jan 2014).

18.

D' Almeida

Kierzek

de Truchis

. Modest public health impact of nontargeted human immunodeficiency virus screening in 29 emergency departments. Arch Intern Med 2012; 172: 12–20.

19.

Haukoos

. Rethinking how we perform HIV testing in the emergency department. Ann Emerg Med 201; 5(8): S160–S163.

20.

Haukoos

. The impact of nontargeted HIV screening in emergency departments and the ongoing need for targeted strategies: comment on “Modest Public Health Impact of Nontargeted Human Immunodeficiency Virus Screening in 29 Emergency Departments.”. Arch Intern Med 2012; 172: 20–2.

21.

Holtgrave

. Costs and consequences of the US Centers for Disease Control and Prevention's recommendations for opt-out HIV testing. PLoS Med 2007; 4: e194–e194.

22.

CDC. HIV in the United States: At a Glancehttp://www.cdc.gov/hiv/statistics/basics/ataglance.html (accessed 11 Jan 2014).

23.

NYC Department of City Planning. Population: Current Population Estimateshttp://www.nyc.gov/html/dcp/html/census/popcur.shtml (accessed 1 Jan 2014).

24.

NYC. HIV/AIDS Surveillance Statisticshttp://www.nyc.gov/html/doh/html/data/epi-reports.shtml (accessed 1 Jan 2014).

25.

NYC Department of Health. HIV/AIDS Annual Surveillance Statisticshttp://www.nyc.gov/html/doh/html/data/hivtables.shtml (accessed 1 Jan 2014).

26.

Fetene

Feleke

. Missed opportunities for earlier HIV testing and diagnosis at the health facilities of Dessie town, North East Ethiopia. BMC Public Health 2010; 10: 362–362.

27.

Champenois

Cousien

Cuzin

. Missed opportunities for HIV testing in newly-HIV-diagnosed patients, a cross sectional study. BMC Infect Dis 2013; 13: 200–200.

28.

Frange

Chaix

M-L

Veber

. Missed opportunities for HIV testing in pregnant women and children living in France. Pediatr Infect Dis J 2014; 33: e60–62.

29.

O'Shea

Ebrahim

Egli

. Late presentation of HIV despite earlier opportunities for detection, experience from an Irish tertiary referral institution. Ir J Med Sci 2013; 182: 389–394.

30.

Kako

Stevens

Mkandawire-Valhmu

. Missed opportunities for early HIV diagnosis: critical Insights from Stories of Kenyan Women Living with HIV. Int J Health Promot Educ 2013; 51(5):

31.

Brawley

MacConnachie

Nandwani

. Missed opportunities for HIV diagnosis: a three-year audit in the West of Scotland. Scott Med J 2013; 58: 173–177.

32.

Deblonde

oker

Hamers

. Barriers to HIV testing in Europe: a systematic review. Eur J Public Health 2010; 20: 422–432.

33.

Ellis

Curtis

Ong

ELC

. British HIV Association (BHIVA), BHIVA Clinical Audit Standards sub-committee. HIV diagnoses and missed opportunities. Results of the British HIV Association (BHIVA) National Audit 2010. Clin Med Lond Engl 2012; 12: 430–434.

34.

Wohlgemut

Lawes

Laing

RBS

. Trends in missed presentations and late HIV diagnosis in a UK teaching hospital: a retrospective comparative cohort study. BMC Infect Dis 2012; 12: 72–72.

35.

Lyons

Lindsell

Wayne

. Comparison of missed opportunities for earlier HIV diagnosis in 3 geographically proximate emergency departments. Ann Emerg Med 2011; 58: S17–S22.e1.

36.

HIV/AIDS Treatment Guidelines|AIDSinfo

http://aidsinfo.nih.gov/guidelines (accessed 26 Aug 2013).

37.

Prejean

Song

Hernandez

. Estimated HIV incidence in the United States, 2006–2009. PloS One 2011; 6: e17502–e17502.

38.

White

DAE

Warren

Scribner

. Missed opportunities for earlier HIV diagnosis in an emergency department despite an HIV screening program. AIDS Patient Care STDs 2009; 23: 245–250.

39.

Klein

Hurley

Merrill

. Review of medical encounters in the 5 years before a diagnosis of HIV-1 infection: implications for early detection. J Acquir Immune Defic Syndr 2003; 32: 143–152.

40.

Dorell

Sutton

Oster

. Missed opportunities for HIV testing in health care settings among young African American men who have sex with men: implications for the HIV epidemic. AIDS Patient Care STDs 2011; 25: 657–664.

41.

Sudarshi

Pao

Murphy

. Missed opportunities for diagnosing primary HIV infection. Sex Transm Infect 2008; 84: 14–16.