Article Text
PDF
Abstract
Objectives To report the implementation and outcomes of a routine opt-out HIV testing policy in an acute medical unit (AMU) of a district general hospital in an area of high diagnosed HIV prevalence.
Methods Since July 2011, all patients aged 16–79 years attending AMU were offered an HIV test as a hospital policy. Consenting and arranging the test was carried out by general medical staff, with training and motivational support by local HIV specialists. A retrospective cross-sectional review was conducted: testing rate and outcomes of those testing HIV seropositive were determined by review of hospital data systems and case notes.
Results Over a 21-month period, there were 12 682 admissions; 4122 (32.5%) had HIV tests. 20 patients (0.48%) were diagnosed with HIV; 17 (85%) of them were new diagnoses. Compared with those patients targeted as a result of clinical suspicion of HIV (n=6), patients who were diagnosed solely due to the scheme (n=14) had higher baseline CD4 counts (median 111 vs 313 cells/mm3; p=0.01). Two patients had renal disease which improved on antiretroviral therapy. Two long-term defaulters to HIV care with very advanced disease have re-engaged resulting in excellent clinical outcomes. 11 patients are now on treatment with undetectable HIV viral loads. One contact tested HIV positive.
Conclusions Our experience shows that routine opt-out testing can be delivered and sustained by general medical staff in an AMU with no money spent other than laboratory processing of the test. We believe that success and sustainability of this policy is due to the high level of commitment from and ownership by the AMU staff, particularly nurses. Ongoing support and motivation from the HIV team has facilitated the delivery of this policy.
- HIV
- Hiv Testing
- Testing
- Policy
- NHS
Statistics from Altmetric.com
Introduction
Despite widespread access to highly active antiretroviral therapy (HAART), HIV infection still causes considerable morbidity and mortality in the UK. This is largely attributable to late diagnosis.1 It is estimated that a quarter of all HIV infections in the UK are undiagnosed,2 hence the drive to increase HIV testing to reduce late presentation.3 Furthermore, individuals who are aware of their HIV infection can modify their risk behaviours4 and access treatment; both of which can help reduce onward HIV transmission.5 ,6 In 2008, the UK national guideline for HIV testing recommended routine HIV testing in all general medical admissions in areas where the diagnosed HIV prevalence in the local population exceeds 2 per 1000 population.3 Numerous UK studies 7–10 show that HIV testing in these settings is feasible and acceptable: the implementation and sustainability of this practice embedded in routine clinical care, beyond research pilots is yet to be demonstrated.
We report our experience of implementing a routine HIV testing policy, delivered by existing acute medical staff, in an National Health Service trust in South London where the local diagnosed HIV prevalence is 4.78 per 1000. Of those diagnosed, 49% are men, 22% are men who have sex with men, 56% are black African and 25% are of white ethnicity.11
Methods
The acute medical unit (AMU) is where patients are assessed prior to admission to specialised wards. A routine opt-out HIV testing policy commenced at Croydon University Hospital, a district general hospital, in July 2011 whereby all patients aged 16–79 years attending the AMU were to have a standard HIV test unless they declined. The assay was funded by a public health budget commissioned as a result of a business case presented to the local Healthcare Commissioning Board. The testing pathway was agreed at a joint meeting attended by senior representatives of AMU, microbiology and HIV departments.
Prior to roll-out, AMU staff were trained by the HIV team at workshops, which focused on offering the test. Similar tutorials were presented to medical trainees; these were subsequently repeated at inductions. Prelaunch publicity comprised posters in the AMU and postgraduate centre, emails to all medical doctors, announcements at grand rounds (medical teaching events) and articles on the Trust intranet. The AMU pro-forma was adapted to allow staff to document prospectively test offer, acceptance and reasons if declined. These data were matched with patient demographics and test results obtained from the medical notes and pathology records, respectively.
Following arrival to the AMU, patients were given a welcome pack, which included information on the HIV testing policy. General medical doctors or nurses from AMU obtained verbal consent for the test, after which an electronic order was made followed by phlebotomy, rather than an HIV request being added to a previously taken serum sample. If the patient had had an HIV test in the previous 12 months the sample was not processed, unless the requester stated a clear reason for repeat testing. Patients were informed that results would be communicated on a ‘No news is good news’ basis. Positive results were communicated from the laboratory to the admitting team.
The HIV team offered support to the medical team in giving the initial reactive HIV test result. On confirmation of the patient's seropositive status, specialist follow-up was arranged, ensuring a seamless link into HIV care. For cases where the patient left the hospital before the initial HIV test was reported, a general medical outpatient appointment was arranged, where the result was given prior to continuing care by the HIV team.
HIV team members of all grades visited AMU every weekday to encourage staff to offer the test and to troubleshoot any problems. From the outset, nurses were more proactive than doctors in applying the policy. Building on their enthusiasm, nurses were encouraged further. By October 2011, AMU visits were withdrawn gradually as nurses took the lead to enforce the policy. A retrospective cross-sectional review was conducted. Testing rate and outcomes of those testing HIV seropositive were determined by review of hospital data systems and case notes.
Results
From 1 July 2011 to 31 March 2013, there were 12 682 eligible AMU admissions, 4122 HIV tests were requested (32.5%) and 20 HIV diagnoses were made (0.48%) (±0.21 95% CI). Demographic data were available for the first 6 months of admissions (n=3709). The median age was 57 years (IQR: 41–70), 50% were women, 54.7% were Caucasian, 14% were Asian, 6% were black African, 6% were black Caribbean and 19.3% were other. There was no difference in the demographics between those who tested for HIV and the whole group of admissions.
In a random sample (n=396), 183 were approached for testing and 154 accepted the test giving an uptake rate of 84% (154/183) (±3.61 95% CI).
The test rate increased from 33.2% in the first 3 months to 41.3% in the second 3 months (p<0.005 χ2). Demographic and clinical data for the 20 patients diagnosed with HIV are shown in table 1. Six were women and median age was 41.5 years (range 31–75). Sixty-five per cent (14/20) had clinical indicator diseases at presentation. Six patients were tested due to a documented clinical suspicion of HIV with median baseline CD4 counts of 111 cells/mm3 (range 1–249). The remaining 14 patients who tested due to the policy had median baseline CD4 counts of 313 cells/mm3 (range 28–1043). Eight of these (57%) presented with clinical indicators of HIV.
Presentation and demographics of all patients diagnosed with HIV on the AMU
One patient had laboratory-confirmed primary HIV infection with a wife 18 weeks pregnant. Now, 2 years later she and her baby remain HIV negative. Two patients were diagnosed with early HIV-associated nephropathy and have since commenced HAART with a subsequent normalisation of renal function. One patient was discharged before his positive HIV result came back. Recall is ongoing. All other patients have been linked into HIV care, 11 of whom have commenced HAART according to British HIV Association guidelines.
It later transpired that three patients already knew their HIV status but did not disclose this to the AMU team. Two of them had defaulted from HIV care elsewhere and have now re-engaged at our HIV clinic. They are now on HAART and are virologically suppressed. The third patient continues to receive HIV care elsewhere.
Regarding the contact tracing outcomes for the 14 patients diagnosed due to the testing policy, there were a total of nine contacts; seven were traceable and six have now taken an HIV test. Four have tested negative. Two have tested positive. One was already aware of her diagnosis; the other had a baseline CD4 155 cells/mm3 (19%) and has since started HAART. Nine offspring were identified, of which four have tested negative and five are living outside the UK.
At £4.98 each, 4122 tests totalled £20 527 equivalent to £1466 for each of the 14 ‘policy attributed’ diagnoses made.
Discussion
Principle findings
Our experience shows that HIV testing in general medical admissions as a policy is acceptable, feasible, sustainable and cost-effective. It has been delivered by existing medical staff alongside their other clinical duties, with time invested by HIV specialists for support, but no money spent by the hospital. Compared with those patients targeted due to clinical suspicion of HIV (n=6), patients who were diagnosed solely due to the scheme (n=14) had significantly higher baseline CD4 counts (median 111 vs 313 cells/mm3; p=0.01). Forty-three per cent (6/14) of the patients diagnosed as a result of the testing policy had no evidence of indicator diseases at presentation. Their HIV diagnosis would have been missed if it were only patients with indicator diseases who had been tested. Conversely eight patients had clinical indicator diseases but were tested due to the policy rather than clinical suspicion. Three of the patients diagnosed due to the policy were over 59 years old supporting our upper age limit of 79.
Comparison with other work
Testing coverage rate was 32.5%, equivalent to what has been achieved in research settings (13–35%).12 We note a substantial increase from our Trust's prepolicy testing rate (0.85%: 9/1047) (Unpublished Audit, R. Hill-Tout). The uptake rate of 84% is comparable with other reports.9 ,10 Positivity rate was 0.48% (20/4122), approximately five times higher than the threshold prevalence for cost-effectiveness quoted in testing guidelines.3
Limitations
Limitations of this study include lack of comprehensive cost-effectiveness analysis and those inherent to retrospective notes review; documentation bias and missing data regarding HIV offer and uptake.
Mechanisms, implications and future work
We believe that ongoing success and sustainability of this testing policy is due to the high level of commitment from and ownership by the AMU staff, particularly nurses. Some of the operational challenges and possible contributory factors to the low coverage rate included short length of stay on AMU, rapid staff turnover and use of bank staff. Routine opt-out HIV testing in AMU in areas of high prevalence can be implemented successfully and sustained with no money spent other than laboratory processing costs. However, clinicians and policy makers may need to consider their local resources and HIV prevalence prior to implementing such a policy. Local pilot studies may be warranted.
Key messages
-
Routine opt-out HIV testing delivered by non-specialist staff can be implemented and sustained in general medical admissions with no money spent other than laboratory processing costs.
-
Nurses play a key role in enhancing test rates: better than if you rely solely on doctors to undertake HIV consenting.
-
Behind the scenes support from HIV specialists is still of use to maintain the momentum of a testing policy.
-
In areas of high HIV prevalence such a programme is effective; diagnosing HIV in those who otherwise may not have tested.
Acknowledgments
The authors thank Andrew Widdowson and Stuart Rees for producing data reports from the hospital electronic record, the staff of AMU for taking on the testing policy and the HIV department for supporting the AMU with training and guidance.
Footnotes
-
Handling editor Jackie A Cassell
-
Contributors DP, SD, RM and AE participated in the conception and implementation of the testing policy. All authors contributed to the training and support of staff carrying out the policy. SD championed the operational delivery of the policy. DP, AE, AB and JS collected and analysed the data and drafted the sections of the manuscript. All authors contributed to review and revision of the manuscript.
-
Funding A fellowship grant was awarded by Gilead Sciences Ltd. This contributed towards administrative costs: data input and printing patient literature. Gilead Sciences had no role in the design, conduction or interpretation of this study. The authors take full responsibility for the content of this manuscript.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.