Feasibility and acceptability of point of care HIV testing in community outreach and GUM drop-in services in the North West of England: A programmatic evaluation

A multidisciplinary, multiagency steering group was established, comprised of clinicians, specialist HIV community nurses, members of the local primary care trust and public health researchers. Minutes, notes and correspondence following meetings were included in the analysis. Since co-authors were also participants in these groups methods that were cognisant of the observer participant approach were used. One author (PM) was an outsider to the process acting more as observer than participant. Others were active participants whose observations were also considered in the analysis. The steering group undertook situational analysis of a number of existing health and social programmes in Liverpool to identify services and programmes likely to interact with individuals at risk of HIV infection (e.g. outreach care programmes for IVDUs and sex workers) and marginalised people who may find it difficult to access health care (e.g. asylum seekers and homeless people). Services meeting these criteria were invited to participate in the POCT programme, representatives of the management were asked to join the steering group and service providers were trained in POCT for HIV. As the services identified offered a variety of health and social programmes, service providers were encouraged to develop strategies to incorporate POCT into their existing services.

To mobilise interest in POCT, an awareness-raising campaign, called Liverpool Gets Tested, was held around the week of World AIDS Day 2009 (1^st December). This included a number of community-based events such as HIV vigils and radio advertising campaigns.

Qualitative methods were chosen to elucidate perceptions and experiences as they allow for a more flexible and in-depth exploration of acceptability of new services. The 25 service providers performing HIV POCT at the 6 sites were invited to participate in three focus group discussions (two with providers from community sites and one with providers from the GUM site). All interviews were recorded and transcribed by the first author. A framework approach to analysis was undertaken. Qualitative analysis software (NVIVO8, QSR Software, Melbourne) was used for coding purposes. Emerging themes were grouped and triangulated with the steering committee members. Transcripts were independently reviewed by a second researcher to confirm themes.

Attendances for POCT at study sites, non-identifiable participant demographic details, and POCT for HIV results (cross-referenced with laboratory confirmatory testing where necessary) were captured by staff in logbooks. Participant demographics and self-reported clinical history were used to code a person as being in a risk exposure category for HIV or otherwise; risk exposure categories included: MSM, IDU, UK African, having bought or sold sex, having been raped, or having an HIV-positive partner. Participants could belong to multiple risk categories. Persons receiving a POCT test for HIV during the week of World AIDS Day were invited to self-complete a questionnaire about their experience of POCT.

POCT for HIV was undertaken using Determine^® HIV-1/2 Ag/Ab Combo (Inverness Medical Inc, UK) and tested on fingerprick blood samples. HIV testing and pre- and post-test counselling was conducted in private rooms within each site by nurses who had received training in test kit use and HIV counselling. Rooms had been previously inspected for suitability based on the presence of adequate lighting; a clean, flat and washable testing surface and accessible hand washing facilities. Participants with a reactive or invalid test (that did not resolve on immediate retesting) had a further sample sent for confirmatory same day fourth generation laboratory ELISA testing. The study nurse or clinician conveyed confirmed positive results to the participant in a follow-up appointment and referral for comprehensive HIV care was made to the HIV clinic at the Royal Liverpool Hospital, with the study clinician acting as a link person for accessing care.

Client characteristics (logbook) and perceptions of POCT for HIV (Liverpool Gets Tested questionnaire) were expressed as proportions and compared between the LCSH and community-based sites using chi-square tests. The Wilcoxon rank sum test was used to compare differences in age distribution between the groups. Hierarchical risk exposure categories (including mutually exclusive categories with a single, or multiple risk exposures) were constructed for participants testing at the two sites and compared using Fisher's exact test. Statistical analysis used STATA v11.2 (Statacorp, College Station, Tx).

Ethical approval was obtained from the Liverpool School of Tropical Medicine, The National Health Service (Liverpool Primary Care Trust and the Royal Liverpool University and Broadgreen Hospitals Trust).

Through the steering group, six sites were identified as offering health and social care services to target groups. These included a hospital-based GUM clinic (the Liverpool Centre for Sexual Health [LCSH]) and five community-based programmes (a drug-users support group; an asylum-seekers health programme; a MSM health and support programme; a travel clinic; and a support programme for homeless people).

Situational analysis of the study sites showed differences between the hospital-based site (LCSH) and the community-based sites that would influence implementation of the POCT protocol (Table 1). LCSH, an outpatient facility of the Royal Liverpool University Hospital, provides sexual health and HIV services to Liverpool and Merseyside, with over 34,000 attendances in 2009. Clients actively self-present for care and LSCH conducts limited community outreach. POCT was conducted in the department as opportunistic screening for individuals presenting for clinical care.

The five community-based sites, although serving different client groups, had a number of common features relevant in planning the introduction of POCT. In particular, they all had existing social support programmes for the client groups they served. Therefore here it was decided to adopt a more community-orientated approach to POCT. This would involve a focus on awareness-raising about HIV and community mobilisation (including organising church vigils and radio campaigns) to encourage uptake of POCT in target and marginalised groups.

Between September 2009 and June 2010, 953 participants underwent POCT for HIV, 556 (58%) at LCSH and 397 (42%) at the community-based sites. During the Liverpool Gets Tested Campaign, 154 people underwent POCT for HIV. Of individuals tested, and where data were recorded, 659/946 (70%) were male and the median age was 29 years (interquartile range (IQR): 23 - 38). A greater proportion of people testing in the community-based sites (74%) were males compared to in LCSH (67%, p = 0.028). However, the proportion of males who self-identified as MSM was higher at LCSH (36%) compared to the community-based sites (44%, p = 0.050)

Complete risk category data was available for 927 (97%) participants, with all missing data from those testing at community-based sites (Table 2). In total, 475 (51%) were found to be in one HIV risk category. A further 41 (4%) participants were in two risk categories and 2 (0.2%) were in three risk categories. The most frequently reported category was MSM (28%), followed by UK African origin (14%) and current or previous intravenous drug use (8%). Reflecting the clients groups served by the two models of POCT (community and hospital-based), participants tested in the community-based sites were more likely to be UK Africans and to report current or previous IDU compared to those tested in LCSH. In contrast, participants tested at the hospital-based LCSH were more likely to have reported having been raped and having been exposed to an HIV-positive partner.

Seventeen confirmed HIV diagnoses were made (prevalence = 1.8%); 14 from LCSH (prevalence = 2.5%) and three from community based sites (prevalence = 0.8%). Ten individuals diagnosed HIV-positive were MSM, four were UK Africans, one reported IDU and one reported having been gang raped. One HIV-positive participant had no identified risk category by our definition. HIV prevalence in those in a risk category was 16/517 (3.1%). All but one are under on-going follow up in clinical settings. The remaining one is an IDU with a chaotic lifestyle who has repeatedly failed to attend appointments but remains in the Liverpool area.

During the Liverpool Gets Tested Campaign, 128/154 (83%) of individuals participating in POCT for HIV completed a questionnaire. Risk exposure among those completing questionnaires categories were similar to that of the overall study population, expecting that no participants who reported having been raped or having an HIV-positive partner completed a questionnaire. Approximately half (65/127, 51%) had never tested for HIV before and when questioned about their experience of POCT 105/125 (84%) preferred POCT to laboratory testing and 115/125 (92%) stated they would recommend POCT to others. There were no differences in reported previous testing for HIV (p = 0.867) or preference for POCT between LCSH and the community-based sites (p = 0.759). However, a greater proportion of participants testing at community-based sites (96%) stated that they would recommend POCT compared to those testing at LCSH (82%, p = 0.015).

Evaluation of minutes of steering group meetings during preparation for and implementation of POCT revealed a number of issues important to others considering multi-disciplinary POCT for HIV programmes. The principal concern raised by service providers was that POCT would impact upon their regular duties. This overlapped with concerns from healthcare commissioners that POCT for HIV could distract from core service provision. Furthermore, questions were raised as to how POCT for HIV would continue to be funded following the completion of the research project.

The competing demands of the health care commissioners to demonstrate value for money of POCT and that of research academics to demonstrate effectiveness of the HIV testing programme required considerable time and mutual co-operation to resolve. Demonstration that new HIV-positive diagnoses were being made amongst marginalised groups such asylum-seekers and drug users helped persuade commissioners of the value of the POCT.

Through analysis of focus group discussions, we identified key themes raised by service-providers in LCSH and the community-based sites about their experiences of participating in the POCT for HIV programme.