UK guidance recommend all acute medical admissions be offered an HIV test. Our aim was to determine whether a dedicated staff member using a multimedia tool, a model found to be effective in the USA, is an acceptable, feasible, and cost-effective model when translated to a UK setting.
Between 14th Jan to 12th May 2010, a Health advisor (HA) approached 19–65 year olds at a central London acute medical admissions unit (AAU) and offered a rapid HIV point of care test (POCT) with the aid of an educational video. Patients with negative results had the option to watch a post-test video providing risk-reduction information. For reactive results the HA arranged a confirmatory test, and ensured linkage into HIV specialist care. Feasibility and acceptability were assessed through surveys and uptake rates. Costs per case of HIV identified were established.
Of the 606 eligible people admitted during the pilot period, 324 (53.5%) could not be approached or testing was deemed inappropriate. In total 23.0% of eligible admissions had an HIV POCT. Of the patients who watched the video and had not recently tested for HIV, 93.6% (131/140) agreed to an HIV test; four further patients had an HIV test but did not watch the video. Three tests (2.2%, 3/135) were reactive and all were confirmed HIV positive on laboratory testing. 97.5% felt HIV testing in this setting was appropriate, and 90.1% liked receiving the information via video. The cost per patient of the intervention was £21.
Universal POCT HIV testing in an acute medical setting, facilitated by an educational video and dedicated staff appears to be acceptable, feasible, effective, and low cost. These findings support the recommendation of HIV testing all admissions to AAU in high prevalence settings, although with the model used a significant proportion remained untested.
Citation: Burns F, Edwards SG, Woods J, Haidari G, Calderon Y, Leider J, et al. (2012) Acceptability and Feasibility of Universal Offer of Rapid Point of Care Testing for HIV in an Acute Admissions Unit: Results of the RAPID Project. PLoS ONE 7(4): e35212. https://doi.org/10.1371/journal.pone.0035212
Editor: Alan Landay, Rush University, United States of America
Received: November 12, 2011; Accepted: March 10, 2012; Published: April 27, 2012
Copyright: © 2012 Burns et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Gilead’s United Kingdom and Ireland Fellowship Programme and Camden’s & Islington’s Provider Services. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This work was undertaken at University College London Hospital/University College London who received a proportion of funding from the Department of Health’s National Institutes of Health Research Biomedical Research Centres funding scheme.
Competing interests: University College London Hospital has received free INSTI-60 second tests from Pasante Healthcare for another testing pilot within the Trust. Gilead funded part of this work through an unrestricted educational grant via their United Kingdom and Ireland Fellowship Programme. JL & YC received funding for HIV testing from Gilead. JC, SE and FB have received funding from various pharmaceutical companies to attend conferences and/or been paid to lecture at educational meetings. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials.
Late presentation to HIV services is the single most preventable cause of HIV related morbidity and mortality . Late presentation also means a person must have been living with undiagnosed HIV infection for a substantial period of time, and people with undiagnosed HIV infection disproportionately contribute to onward transmission of the infection . For these reasons, and for the associated increase in healthcare costs that late presentation brings , HIV prevention efforts have increasingly focused on improving opportunities to have an HIV test so as to reduce both late presentation and undiagnosed HIV infection –.
The Jacobi Medical Center and North Bronx Healthcare Network (NBHN), New York, developed Project B.R.I.E.F. (Behavior intervention, Rapid HIV test, Innovative video, Efficient cost and health care savings, Facilitated seamless linkage to outpatient HIV care). This program uses a “public health advocate”(PHA) to recruit stable patients attending an inner-city emergency department. It uses a model of universal testing with no specific population targeted. A multimedia tool (tablet PC) is used to deliver a validated video for HIV pre-test counselling (the video lasts ∼90 seconds and is available in 2 languages), followed by a rapid POCT. A post-test video, viewed while the individual waits for the test results, delivers risk reduction counselling and education. The educational videos are as effective as in-person counselling in conveying information related to testing . Basic demographic and risk factor data are collected through this tool using touch screen technology. Using this model, of 7109 eligible patients who watched the video, 87% (6218) were tested, identifying 57 new infections .
The publication of the National guidelines on HIV testing  in 2008 led to a number of initiatives to assess the feasibility, acceptability and cost effectiveness of new models of delivery for HIV testing. Our aim was to determine whether a model of care utilising a multimedia tool and dedicated staff and found to be effective in an emergency medical setting in New York, is an acceptable, feasible and cost effective model in reducing late presentation of HIV infection when translated to a UK setting.
The local Research Ethics Committee, part of the National Research Ethics Service, waived requiring formal ethical approval under SL24 Project not considered to be research, version 4.0 April 2009, as they regarded the pilot as service evaluation.
Data was collected on all new admissions to an acute admission unit (AAU) in Central London over a 16 week period (15th Jan to 11th May 2010 inclusive). Adults aged 19–65 in a stable clinical condition were eligible for inclusion in the HIV testing pilot. Patients who were only on the AAU during the weekend when the HIV testing service was not available were excluded from this analysis.
The service model employed consisted of a health advisor (HA) approaching all new stable admissions, and offering HIV testing with the aid of an educational video available in up to 4 languages. If patient accepted, a finger prick rapid HIV point of care test (POCT) was performed using the EU approved INSTI™ test (sensitivity and specificity reported elsewhere) . If the result was HIV negative the patient had the option of watching a post-test video providing risk-reduction information. If the result was reactive the HA explained the need for a confirmatory test, arranged the test and urgent follow up with the HIV service. All patients watching either video completed a questionnaire designed to evaluate patient satisfaction and collect process evaluation data (demographic and behavioural profiles). The questionnaire was delivered electronically with touch-screen technology via the tablet PC that patients used to watch the videos.
The video scripts were adapted from those used in Project BRIEF to suit a UK context and to meet BHIVA guidelines . They were pretested in two focus groups with service providers and users. At the time of the pilot only an audio English version was available but now versions exist with English, French, Polish, and Spanish subtitles.
The tablet PC and all equipment necessary for the POCT tests were located on a small portable trolley that was wheeled to the bedside. Disposable headsets were used so as not to disturb other patients.
Following the completion of the project all staff who worked on the unit during the pilot period were surveyed about their experiences. Surveys were distributed at meetings for those still working on the unit and via email to those now located elsewhere.
Standard statistical tests, e.g. Chi2 test and Student’s t-test, were used to examine associations between variables.
We directly calculated the incremental cost of the educational video intervention versus treatment as usual from a National Health Service (NHS) perspective. The cost components included in the analysis were video equipment costs, cost of disposable equipment and staff costs incurred through training, delivering the intervention and post-test counselling. Resource use data for each cost component were collected in the study. We applied unit costs from market prices and published sources . We conservatively assumed that the lifetime of the video equipment was approximately 1000 patients. The majority of the costs incurred were staff costs. We assumed in our main analysis that these would be incurred by a Health Advisor (Band 7), based on 3 tests per hour. In sensitivity analyses we explored the impact of using different staff and increasing the number of tests per hour.
During the study period there were 606 eligible (19 to 65 years of age inclusive and inpatient on weekday) admissions to the AAU, representing 602 individuals. As none of the repeat attendees had an HIV test on their first visit both visits are included in subsequent analyses. Three quarters (456/606, 75.3%) of all eligible admissions were approached to participate in the study. There were no significant differences in the gender, age, ethnicity, presentation pattern or length of stay between those approached and not approached (table 1).
Despite often multiple attempts, over half (53.5%) of approaches failed to encounter or engage the patient (table 2). Of the 282 patients whom were asked if they would be involved in the pilot project, 153 (54.3%) agreed. On introduction of video four patients asked or agreed to have an HIV test but did not want to watch the video, and five disclosed that they had recently tested for HIV and therefore withdrew from further involvement. Following the video a further eleven patients declined to test: four had tested within the past three months; two had never been sexually active; two declined because of communication difficulties; one wanted to test in an anonymous environment and was referred to the local sexual health clinic; one became unwell during the video; and one declined. In all, of the 140 patients who watched the video and had not tested for HIV in the preceding three months, 93.6% (131/140) agreed to an HIV test. All patients tested received their results at the time of testing.
Older people (aged 40 years or more) were less likely to accept to watch the video than younger patients (43.4% vs. 64.2%, p = 0.001), however if they did so uptake of the test did not differ by age (71.8% vs. 76.9%, p = 0.476). There was no difference in the uptake of the video or HIV test by gender.
In total 23.0% of eligible admissions to AAU during the pilot period had a POCT HIV test, and 25.7% left AAU knowing their HIV status having tested on that admission or within the preceding three months, or previously been diagnosed HIV positive. Three tests (2.2%, 3/135) were reactive on POCT and all were confirmed HIV positive on further laboratory testing.
The three patients diagnosed HIV positive were a 48 year old British man with pneumonia, presumed Pneumocystis jirovecii pneumonia (PCP), CD4 20; a 42 year old Nigerian woman admitted with bacterial pneumonia, CD4 40; and a 60 year old British man with rectal bleeding on warfarin, CD4 590. All three patients were seen by specialist HIV services whilst an inpatient and remain engaged with HIV services 12 months on. Only one of the three had previously tested for HIV, the 48 year old British man having tested over 5 years previously.
The majority of RAPID participants (patients who watched the video and completed the survey) were male (58.6%), and the median age was 38.5 years. Over half (51.9%) resided in the hospital catchment area (local boroughs of Camden or Islington), and 85.5% were from within London. Over two fifths (42.8%) of participants were born abroad: 19 (12.5%) in Europe, 17 (11.2%) in Africa (9, 5.9% black African) and 15 (9.9%) in Asia or the Indian sub-continent. During the pilot the video was only available in spoken English (subsequently versions with English, French, Polish and Spanish subtitles have been developed), despite this 87.5% of patients stated the video was in their preferred language.
Forty percent (61/152) of participants had previously had an HIV test, however only 22 (14.5%) had tested within the past 12 months and 19 (12.5%) had tested more than 5 years previously. Five participants who had previously tested (14.8%, 5/61) reported they never received the results of their last HIV test. Almost 20% of participants reported behaviour associated with increased risk for HIV (table 3). Prior HIV testing was more prevalent in those reporting an HIV risk behaviour than those who did not (75.0% vs. 32.8%, p<0.001).
The overwhelming majority (97.5%) of participants of the RAPID pilot thought POCT HIV testing in the AAU was a good idea, and 96.7% thought rapid HIV testing appropriate in this setting. Almost all (90.1%, 137/152) participants liked receiving information via video, and the majority (80.3%, 133/152) felt the short video answered their questions about HIV testing. The main reason for patients not liking the use of video was because they would prefer face to face interaction (11/14); two patients also mentioned that they were not able to ask questions they wanted to, and another two felt the leaflet introducing the project had already provided sufficient information. Participants were asked if the Health advisor running the pilot made it easy to get tested: 82.9% responded ‘yes, very’, 10.5% ‘yes, a little’, and 5.9% ‘not sure’.
Finally participants were asked how they would prefer rapid HIV testing to be offered: 2.6% preferred video alone, 15.8% a health advisor alone, 9.2% a combination of health advisor and leaflet, 46.1% the combination of video and health advisor, and 25.7% a combination of health advisor, video and leaflet.
Acceptability to Staff
Responses were obtained from 61.5% (88/143) of clinical staff working on AAU during the pilot; the response rate was lower amongst doctors of all grades (50%), than among nurses and health care assistants (74.1%). No staff felt the service had disrupted their job in any way, and all staff felt the service should be continued. 92% of doctors believed that more of their own patients were now tested for HIV (table 4), and no doctors felt the service made them less likely to offer a test; with three-quarters of doctors believing the service increased the likelihood of them requesting an HIV test either directly or via the service.
Regular POCT testing on AAU continued until June 25th, and the number of standard laboratory HIV test requests in AAU in the five months preceding and during pilot did not significantly differ (data not shown).
The additional cost of the equipment required for the educational video was £1709 (Table 5). The incremental cost of the education video intervention per patient was £21 (Table 6). The largest component of the cost was the staff cost to run the video and test and associated administration (49% of the total incremental cost). The cost per case identified was £1,083. If the costs of disposable equipment were excluded on the basis that these would have been incurred in any case, then the incremental cost of the education video per patient fell from £21 to £15. If the service was provided by a Nurse Band 5 rather than a Health advisor Band 7 the cost per patient fell from £21 to £18. If it was provided by a Healthcare Assistant they fell to £14. If 6 rather than 3 tests were undertaken per hour then the costs per patient were £16, £14 and £12, depending on whether the staff member involved was a Health advisor Band 7, Nurse Band 5 or Healthcare Assistant.
Routine HIV point of care testing in the AAU setting was acceptable to patients. It was successful in identifying cases of HIV and demonstrates the potential for earlier diagnosis in screening those without indicator diseases. Using a service model with a dedicated staff member compared to one where HIV testing is embedded within routine clinical practice undoubtedly increased costs, however the cost per case identified still compares favourably with other screening programmes in the UK . The provision of a dedicated staff member also ensured staff acceptability and no disruption to current services.
For this pilot we deliberately used a senior Health Advisor with a lot of experience in delivering HIV results however with appropriate training and support a more junior staff member could run the service. In project BRIEF in New York, this role is fulfilled by ‘Public Health Advocates’, who usually have no formal clinical training other than the two weeks of training they get once employed on the project.
The use of digital media (the video) ensured consistent messaging, and that information was conveyed in an easy manner for patients with health literacy or literacy issues. Digital media has the ability to overcome linguistic issues, to be flexible around patient care, and can be delivered on sustainable system wide tools, for example patient television. Furthermore the use of video was liked by patients although the survey data suggests that face to face contact time remains important. Unfortunately different models of service delivery, such as video or health advisor alone, were not able to be assessed in the current study.
For the purposes of this paper costings are based on the first 1000 patients seen however the video equipment would be expected to last well beyond the first 1000 patients and subsequent costs would reduce accordingly. 22% of all patients aged 19–65 years were admitted to the AAU and subsequently discharged between 5 pm Friday and 9 am Monday and thus excluded from this study. These patients were likely to be younger (mean age 40 versus 44 years) but there was no gender difference compared with patients on AAU during testing hours. The authors would expect more alcohol related admissions over weekends but whether this directly equates with HIV risk is unknown.
The main criticism of this model could be that it fails to embed HIV testing within routine clinical practice, often referred to as ‘normalising HIV’ . This is a concern that the authors share, however without the will of clinical colleagues in other specialties, sustained routine HIV testing within general medical practice in the UK currently appears to elude us; the notable exception to this of course being the hugely successful universal antenatal screening programme . Implementation of the HIV antenatal screening programme was supported by specific national health policy . While guidelines have been published recommending expansion of HIV testing opportunities to setting such as AAUs these fall short of policy recommendations.
A further criticism could be that two of the three cases identified were likely to have been detected through the already locally established practice of targeted testing of high risk individuals as defined by indicator disease or risk group , and that expanding practice to universal testing is unnecessary. The authors would certainly like to believe that the two cases with indicator diseases would have been identified without the RAPID pilot in place, unfortunately published data suggests that this may not necessarily have occurred , . Earlier diagnosis is increasingly acknowledged as fundamental in prevention of HIV related morbidity and mortality, and in the prevention of onward transmission of infection. Testing based on indicator diseases, as found in the current study, is unlikely to identify people in the early stages of HIV infection. The third case, the 60 year-old gentleman admitted with rectal bleeding, reported no high risk behavior and to our knowledge had never accessed sexual health services; without universal testing his HIV would almost certainly have remained undetected until presenting at some later date with advanced disease.
No difference was found between those approached and not approached in terms of gender, ethnicity, patient stay, or indicator disease status, suggesting the pilot used a truly non targeted approach. Generalisability to other clinical settings however is hindered by the highly specific population and setting of the study. HIV case-finding outside of the London AAU setting is likely to influence the cost-effectiveness of the model.
For any screening procedure to be viable it must fulfill certain criteria, often called the Wilson-Jungner criteria . The criteria are listed in short below followed by findings based on current study.Wilson-Jungner Screening Criteria and HIV testing in AAUs.
- The condition that is being sought must be sufficiently common in the group being screened to make screening worthwhile.
- Screening will lead to earlier detection of a treatable disease so that outcome is significantly improved.
- The screening technique has acceptable costs per case identified and the procedure is acceptable to both patients and staff.
- There will be a high enough uptake to make the procedure valid.
- There will be high specificity (low rate of false positives) and a very high sensitivity (very low rate of false negatives).
- Investigation and management of positive results will not overburden the system.
It is estimated that prevention of one new HIV infection in the UK saves between £280,000 and £360,000 in direct lifetime healthcare costs . Thirty five Primary care Trusts in England have a prevalence of diagnosed HIV greater than 2 per 1,000 adult population. Whilst failing to embed HIV testing within routine clinical practice, utilization of a model of universal POCT HIV testing in an acute medical setting, facilitated by an educational video and dedicated staff may play an important role in the transition to normalization of HIV testing, as this model appears to be acceptable to both staff and patients, feasible, effective, and cost-effective.
With minimal staff training this model could also be adapted to one of universal POCT testing within routine clinical care. A clearly identified pathway to link those with reactive tests into specialist care for confirmatory testing, post test counselling, and linkage into care should ideally support any such initiative, ideally through the provision of a HIV liaison nurse/health advisor.
We are especially grateful to all the staff and patients on the Acute Admission Unit at UCLH.
Conceived and designed the experiments: FB MB SE JC YC JL RT. Performed the experiments: FB MB SE JW. Analyzed the data: FB GH MB SM. Wrote the paper: FB MB SE JC YC JL RT JW GH SM.
- 1. BHIVA Audit & Standards Sub-Committee (2006) BHIVA Audit & Standards Sub-Committee2006Mortality audit. BHIVA website http://www.bhiva.org/NationalAuditReports.aspx. Mortality audit. BHIVA website http://www.bhiva.org/NationalAuditReports.aspx.
- 2. Marks G, Crepaz N, Janssen RS (2006) Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS 20: 1447–1450.G. MarksN. CrepazRS Janssen2006Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA.AIDS2014471450
- 3. Krentz HB, Auld MC, Gill MJ (2004) The high cost of medical care for patients who present late (CD4 <200 cells/microL) with HIV infection. HIV Med 5: 93–98.HB KrentzMC AuldMJ Gill2004The high cost of medical care for patients who present late (CD4 <200 cells/microL) with HIV infection.HIV Med59398
- 4. National Institute for Health & Clinical Excellence (2011) Increasing the uptake of HIV testing to reduce undiagnosed infection and prevent transmission among black African communities living in England. NICE public health guidance 33. 2011. National Institute of Health & Clinical Excellence. National Institute for Health & Clinical Excellence2011Increasing the uptake of HIV testing to reduce undiagnosed infection and prevent transmission among black African communities living in England.NICE public health guidance 33. 2011. National Institute of Health & Clinical Excellence
- 5. National Institute for Health & Clinical Excellence (2011) Increasing the uptake of HIV testing among men who have sex with men. NICE public health guidance 34. National Institute for Health & Clinical Excellence. NICE public health guidance. National Institute for Health & Clinical Excellence2011Increasing the uptake of HIV testing among men who have sex with men.NICE public health guidance 34. National Institute for Health & Clinical Excellence. NICE public health guidance
- 6. Branson BM, Handsfield HH, Lampe MA, Janssen RS, Taylor AW, et al. (2006) Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 55: 1–17.BM BransonHH HandsfieldMA LampeRS JanssenAW Taylor2006Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings.MMWR Recomm Rep55117
- 7. Calderon Y, Haughey M, Bijur PE, Leider J, Moreno-Walton L, et al. (2006) An educational HIV pretest counseling video program for off-hours testing in the emergency department. Ann Emerg Med 48: 21–27.Y. CalderonM. HaugheyPE BijurJ. LeiderL. Moreno-Walton2006An educational HIV pretest counseling video program for off-hours testing in the emergency department.Ann Emerg Med482127
- 8. Calderon Y, Leider J, Hailpen S, H P, Chin R (2009) High-Volume Rapid HIV Testing in an Urban Emergency Department. AIDS Patient Care STDS 23: 749–755.Y. CalderonJ. LeiderS. HailpenP. HR. Chin2009High-Volume Rapid HIV Testing in an Urban Emergency Department.AIDS Patient Care STDS23749755
- 9. British HIV Association, British Association of Sexual Health and HIV, British Infection Society (2008) UK National Guidelines for HIV Testing 2008. British HIV Association. Association British HIVBritish Association of Sexual Health and HIV, British Infection Society2008UK National Guidelines for HIV Testing 2008.British HIV Association
- 10. Fonseca K, DiFrancesco L, Galli R, Hogg R, Schechter M, et al. (2004) Results from a Multi-Centre Canadian Clinical Trial of a Rapid HIV Antibody Test for Use in Point-of-Care, Clinical and Laboratory Settings. XV International AIDS Conference. 2004. Bangkok, IAS. K. FonsecaL. DiFrancescoR. GalliR. HoggM. Schechter2004Results from a Multi-Centre Canadian Clinical Trial of a Rapid HIV Antibody Test for Use in Point-of-Care, Clinical and Laboratory Settings.XV International AIDS Conference. 2004. Bangkok, IAS
- 11. Curtis L (2010) Unit costs of Health and Social Care 2010. Personal Social Services Research Unit, University of Kent. L. Curtis2010Unit costs of Health and Social Care 2010.Personal Social Services Research Unit, University of Kent
- 12. Streetly A (2002) Policy decision for implementing neonatal screening for sickle cell disease. Available: sct.screening.nhs.uk/getdata.php?id = 10621. Accessed 2012 April 3: A. Streetly2002Policy decision for implementing neonatal screening for sickle cell disease. Available: sct.screening.nhs.uk/getdata.php?id = 10621.Accessed 2012 April3
- 13. De Cock KM, Johnson AM (1998) From exceptionalism to normalisation: a reappraisal of attitudes and practice around HIV testing. BMJ 316: 290–293.KM De CockAM Johnson1998From exceptionalism to normalisation: a reappraisal of attitudes and practice around HIV testing.BMJ316290293
- 14. Simpson WM, Johnstone FD, Boyd FM, Goldberg DJ, Hart GJ, et al. (1998) Uptake and acceptability of antenatal HIV testing: randomised controlled trial of different methods of offering the test. BMJ 316: 262–267.WM SimpsonFD JohnstoneFM BoydDJ GoldbergGJ Hart1998Uptake and acceptability of antenatal HIV testing: randomised controlled trial of different methods of offering the test.BMJ316262267
- 15. Department of Health (2003) Screening for infectious diseases in pregnancy: standards to support the UK antenatal screening programme. London. Department of Health2003Screening for infectious diseases in pregnancy: standards to support the UK antenatal screening programme.London
- 16. Peck L, Ferenczi E, Burns F, Cosgrove C, Brown M (2010) Barriers to targeted HIV testing on an acute admissions unit: evaluation of the UK guideline. QJM 103: 147–151.L. PeckE. FerencziF. BurnsC. CosgroveM. Brown2010Barriers to targeted HIV testing on an acute admissions unit: evaluation of the UK guideline.QJM103147151
- 17. Sullivan AK, Curtis H, Sabin CA, Johnson MA (2005) Newly diagnosed HIV infections: review in UK and Ireland. BMJ 330: 1301–1302.AK SullivanH. CurtisCA SabinMA Johnson2005Newly diagnosed HIV infections: review in UK and Ireland.BMJ33013011302
- 18. Wilson JMG, Jungner G (1968) Principles and Practice of Screening for Disease. WHO Chronicle 22(11): 473.JMG WilsonG. Jungner1968Principles and Practice of Screening for Disease.WHO Chronicle2211473
- 19. Chadborn TR, Delpech VC, Sabin CA, Sinka K, Evans BG (2006) The late diagnosis and consequent short-term mortality of HIV-infected heterosexuals (England and Wales, 2000–2004). AIDS 20: 2371–2379.TR ChadbornVC DelpechCA SabinK. SinkaBG Evans2006The late diagnosis and consequent short-term mortality of HIV-infected heterosexuals (England and Wales, 2000–2004).AIDS2023712379
- 20. Chadborn TR, Baster K, Delpech VC, Sabin CA, Sinka K, et al. (2005) No time to wait: how many HIV-infected homosexual men are diagnosed late and consequently die? (England and Wales, 1993–2002). AIDS 19: 513–520.TR ChadbornK. BasterVC DelpechCA SabinK. Sinka2005No time to wait: how many HIV-infected homosexual men are diagnosed late and consequently die? (England and Wales, 1993–2002).AIDS19513520
- 21. Health Protection Agency (2009) HIV in the United Kingdom: 2009 Report. London, Health Protection Agency. Health Protection Agency2009HIV in the United Kingdom: 2009 Report.London, Health Protection Agency