The contribution of HIV point-of-care tests in early HIV diagnosis: community-based HIV testing monitoring in Catalonia, 1995 to 2018

Background Community-based HIV testing services combined with the use of point-of-care tests (POCT) have the potential to improve early diagnosis through increasing availability, accessibility and uptake of HIV testing. Aim To describe community-based HIV testing activity in Catalonia, Spain, from 1995 to 2018, and to evaluate the impact of HIV POCT on the HIV continuum of care. Methods A community-based network of voluntary counselling and testing services in Catalonia, Spain has been collecting systematic data on activity, process and results since 1995. A descriptive analysis was performed on pooled data, describing the data in terms of people tested and reactive screening test results. Results Between 1995 and 2018, 125,876 HIV tests were performed (2.1% reactive). Since the introduction of HIV POCT in 2007, a large increase in the number of tests performed was observed, reaching 14,537 tests alone in 2018 (1.3% reactive). Men who have sex with men (MSM), as a proportion of all people tested, has increased greatly over time reaching 74.7% in 2018. The highest percentage of reactive tests was found in people who inject drugs followed by MSM. The contribution of community-based HIV testing to the overall total notified cases in the Catalonia HIV registry has gradually increased, reaching 37.9% in 2018, and 70% of all MSM cases. In 2018, the percentage of individuals with a reactive screening test who were linked to care was 89.0%. Conclusion Our study reinforces the important role that community-based HIV POCT has on the diagnosis of HIV in key populations.


Introduction
In recent years, efforts to reach the 90-90-90 targets (90% of all people living with HIV knowing their HIV status, 90% of all people diagnosed with HIV receiving antiretroviral therapy and 90% of all people receiving antiretroviral therapy having viral suppression) advocated by the Joint United Nations Programme on HIV/ AIDS (UNAIDS), have led to an improvement in accessibility and coverage of testing programmes. This, in turn, has reduced the number of people living with undiagnosed HIV infection and increased early diagnoses [1]. Monitoring and evaluation (M&E) is an essential component of any effective testing programme. While strategic information should guide the design of testing initiatives, M&E permits continuous evaluation of targets and programme effectiveness, efficiency and impact. Such data can prove invaluable in planning improvements [2].
Catalonia is an autonomous community located in the north-east of Spain. In 2018, it had a population of 7,543,825 inhabitants. This region has a low-level HIV epidemic, where high levels of infection are found only in specific groups, particularly men who have sex with men (MSM). As of 31 December 2017, Catalonia had a rate of 8.1 HIV diagnoses per 100,000 inhabitants, with 53.6% of all diagnoses in 2017 attributed to MSM [3].
In Catalonia, HIV testing M&E forms part of the Integrated AIDS/HIV/STI Surveillance System of Catalonia (SIVES) [4] and is based on two main sources of information: (i) the network of public hospital laboratories, primary healthcare centre laboratories and private laboratories (HIVLABCAT), which have voluntarily reported diagnostic HIV testing and results since 1992; and (ii) the network of community-based voluntary counselling and testing (CBVCT) services, which has offered free, anonymous, voluntary and confidential counselling and testing since 1995 [5]. This report will focus on the data collected by the CBVCT network.
CBVCT services are considered an effective strategy for HIV testing, especially for key populations [6,7], and have expanded in the European Union/European Economic Area (EU/EEA) since 2010 through a variety of service delivery models [8]. This strategy has been proven to increase the availability, accessibility and uptake of HIV testing in order to reduce the number of people who do not know their HIV status or who are diagnosed late [9], impacting the first 90 target set by UNAIDS [10]. In addition, this strategy increases the proportion of first-time testers, increases the proportion of participants who undertook follow-up CD4 tests after diagnosis, detects patients at an earlier stage of infection, increases the number of new HIV diagnoses, and potentially reduces the stigma and discrimination faced by key populations [6].
A systematic review found that the use of HIV pointof-care tests (POCT) as part of CBVCT interventions, combined with behavioural interventions either at individual or community level, has the potential for enormous impact on the HIV epidemic [11]. Scaling up the CBVCT service model was thought to increase the likelihood of achieving the 90-90-90 target by 2020 [12], but the scale up in Europe has been impacted by limited funding, poor integration with national HIV programmes and regulatory barriers. There is a need for guidance to address these implementation challenges, including M&E, and a need to assist countries in developing, implementing and evaluating national policies [13].
Community-based testing started in Catalonia with only a few sites offering traditional testing, where a nurse was required to perform venepuncture and send the blood sample to a laboratory. Traditional testing was replaced with HIV POCT in 2007, which allowed the expansion of testing programmes in the community. Since 2007, more sites have been offering HIV POCT, and the number of tests performed has increased exponentially [5].
Catalonia has experience in the scaling up of CBVCT interventions using HIV POCT with linkage to care, support and treatment services, within a solid M&E framework. Here, our aims are to describe HIV testing activity among those CBVCT services participating in the DEVO (an abbreviation of 'voluntary detection' in Catalan) network from 1995 to 2018 in order to evaluate HIV POCT contribution in the HIV continuum of care.

Settings
In 1995, the Catalan Health Department (currently, the Public Health Agency of Catalonia, ASPCAT) funded a network of CBVCT services to offer free, voluntary and confidential HIV testing in the region. The purpose of the DEVO network was to complement existing facility-based HIV testing. The DEVO network has since expanded from four CBVCT services in 1995 to the current 12 (becoming six organizations in 2001, seven in 2003, eight in 2004, 10 in 2008 and 12 in 2010), mainly operated by NGOs and serving the general population or, in some cases, key populations: MSM, sex workers (SW), young people (under 30 years old), and people who inject drugs (PWID). The participating organisations of the network are all CBVCT services providing HIV testing by trained lay providers through community and outreach services. In addition to providing HIV testing, most organisations perform syphilis and hepatitis C testing and additional HIV prevention activities. One of the organisations also offers other STI tests. Peers and other lay providers have been trained to perform and interpret rapid diagnostic tests with finger-prick blood samples.
Every person who receives a preliminary reactive test is referred to a laboratory or to an infectious disease specialist who conducts a confirmatory test. The diagnosed clients are then linked with appropriate specialist services.

Data collection
The DEVO network has been collecting systematically standardized data on activity, process and results since 1995. Since 2014, the DEVO network has formed part of the community-based testing (COBATEST) network, a European network of CBVCT services based on the DEVO network experience [8,14]. Since then, all except one of the CBVCT services use the COBATEST data collection tool (which is based on the DEVO data collection tool) and COBATEST web-based data entry tool through which data can be extracted and analysed in collaboration with the Centre for Epidemiological Studies on Sexually Transmitted Infections and HIV/ AIDS of Catalonia (CEEISCAT) as part of the Public Health Agency of Catalonia (ASPCAT). One of the CBVCT services uses their own data collection tool, and shares the minimum agreed data with CEEISCAT. For monitoring and evaluation purposes, the network currently uses the standardised core indicators defined in the COBATEST network [15], aligned with UNAIDS, World Health Organization (WHO) and European Centre for Disease Prevention and Control (ECDC) recommendations [16][17][18].
Data collected in the DEVO network include basic demographic information on the tester, test location, testing history, risk behaviour and results of HIV, syphilis and hepatitis C testing. Since 2014, services in the DEVO network have used a unique identifier for each client, ensuring anonymity while allowing the identification of repeat testers and recording the correct number of individuals tested.

Data analysis
The descriptive analysis was performed on pooled data from 1995 to 2018 and included: (i) the whole time period 1995-2018; (ii) each year individually; (iii) the percentage of people tested distributed by gender,         The lines in Figure 4 chart the evolution of the percentage of reactive screening tests per year by transmission group and were smoothed using the centred moving average method. Using this method, data points were modified four times, each time the average of raw observations at a given point in time was calculated using that point, the one immediately prior and the one immediately after. This method allows for smoothing out short-term fluctuations and highlights long-term trends or cycles [22]. To test trends in Figures 1 and 4, a Pearson's chi-squared test was used.
Linkage to care was defined as 'entry into healthcare or follow-up by a HIV specialist or a HIV unit after a reactive or confirmatory HIV test at a community testing facility' according the definition established in the Euro HIV EDAT project, co-funded by the European Commission [23], and all linkage to care information was collected from patient feedback.
In order to evaluate the contribution of community testing to the total number of diagnosed cases in Catalonia, data from the DEVO network and the Catalonia HIV registry were triangulated, considering that reactive tests detected in the DEVO network were linked to care and therefore were noted in the Catalonia HIV registry. The percentage of HIV cases diagnosed in the community and registered in the Catalonia HIV registry was calculated from 2001 to 2017, the period where Catalonia HIV registry data were available.
All percentages were calculated excluding missing values (which represented less than 5%). A p value below 0.001 was considered for statistical significance. Data analysis was performed using PASW Statistics for Windows, version 18.0 (SPSS Inc., Chicago, United States).

Ethical statement
Ethical approval was not needed at the beginning of the project in 1995, as no biological samples were preserved for the study and data collected from the clients were anonymous and part of the routine services   of CBVCT centres. Since the introduction of POCT, each client taking a POCT gave informed consent stating that they understood the POCT was not a diagnostic test. Since 2014 when the DEVO network joined the COBATEST network, the CBVCT services signed an agreement ensuring they fulfilled the General Data Protection Regulation, where each client has to sign an informed consent explaining the use of the data collected.

Results
Between 1995 Figure 4). Nevertheless, in recent years the reactivity rate in the PWID group has gradually decreased. This decrease is not statistically significant, due to the low number of PWID tested. The reactivity rate in MSM plus MSW has shown a statistically significant decrease (p < 0.001), especially in the past 10 years, reaching 2.2% (181/8,209) in 2018. For the rest of the groups, no significant trend was observed. Figure 5 shows the increase of the contribution of HIV POCT in the community to the overall total number of cases registered in the Catalonia HIV registry. The percentage of positive cases in the Catalonia HIV registry which were first detected in the DEVO network has gradually increased, from 4.5% (34/763)

Discussion
This study shows the contribution of community-based HIV POCT in improving early HIV diagnosis in Catalonia over time, especially among key populations, and demonstrates that the collected data are an important source of strategic information to be included into the Integrated AIDS/HIV/STI Surveillance System of Catalonia (SIVES).
In Catalonia, community-based HIV testing has been monitored and has formed part of HIV Surveillance since 1995. The DEVO Network has made it possible to collect standardised data on each person tested in CBVCT services. The collected data complement strategic information on key populations and thus make it possible to improve HIV prevention strategies aimed at these key populations. The continual monitoring performed by the DEVO Network has improved public health decision-making at the Public Health Agency of Catalonia by detecting changes in HIV testing uptake, in HIV tester profiles and in HIV test-seeking behaviours [5].
The DEVO Network succeeded in scaling up HIV testing among key populations, with the number of tests performed in the year following the implementation of POCT increasing by 103% from the previous year [5].
HIV POCT have the potential to increase the number of people who know their HIV status [24]. The POCT that meet the WHO's ASSURED (affordability, sensitivity, specificity, user-friendly, rapid and robust, equipmentfree and deliverable) criteria [25] follow a simple procedure involving a limited number of steps and are equipment-free, ensuring they can be performed outside traditional laboratory settings by staff with no formal laboratory training [24]. Additionally, both providers and clients prefer rapid tests over traditional tests [26,27]. Several studies have shown the efficacy of CBVCT strategies using HIV POCT to improve HIV testing uptake in populations at higher risk of exposure to HIV [6,7,9]. The DEVO Network has shown to be successful in providing testing to at-risk populations.
In the period 1995-2018, 57.2% of tests were performed on MSM, 7.6% on SW (FSW plus MSW), 2.6% on PWID, 1.5% on transgender people and 45.2% on foreign nationals (including migrant population). A recent study showed that in Catalonia, 12.3% of those living with HIV were still undiagnosed, and this proportion was higher in migrants [28]. Therefore, in Catalonia, providing access to HIV POCT in the community is  important, especially for populations facing barriers to accessing the healthcare system, such as the migrant population.
The low number of PWID tested in the DEVO network can be explained by the fact that most PWID are tested in harm reduction centres. In Catalonia there is a network of harm reduction programmes run by mobile units, street teams or facility-based centres. The facility-based centres are located in areas of drug trafficking and drug consumption, or in drug treatment clinics [29].
Linkage to care and treatment for those with a reactive test in the DEVO network is high (89%). A recent systematic review and meta-analysis of studies in the WHO European Region [30] showed a pooled estimate of 85% (95% CI: 75-93) of people with reactive tests linked to care within 3 months. Linkage of those with a reactive test to appropriate specialist services is a key step in the HIV continuum of care, as immediate initiation of treatment has substantial benefits in reducing the risk of patient morbidity, as well as reducing onward transmission [30].
In the past 10 years, a statistically significant decreasing trend has been observed in the percentage of MSM with a HIV reactive test. This trend could be explained by the success of different strategies of combined prevention in this key population, including increased testing frequency and earlier initiation of HIV treatment. BCN Checkpoint (the CBVCT service with the largest HIV testing activity, particularly among MSM) has gone further to promote earlier initiation of HIV treatment by introducing qualitative PCR POCT for the detection of acute HIV infection [31]. This, coupled with their preexposure prophylaxis (PrEP) service in the framework of research studies, has broadened the portfolio of preventive services available to users of the Checkpoint. In England, the incidence of new HIV infections in MSM attending sexual health clinics fell by 55% in 2016 and 2017 [32,33] The longstanding experience of the DEVO Network and its results has been used as a basis for establishing the COBATEST Network, a European network of CBVCT services that share standardised data [8,14].
There are a number of limitations to this study. Firstly, the disaggregated data presented in the Table from 1995 to 2013 refers to the number of tests performed, while data from 2014 to 2018 refer to the total number of people tested. This is due to improvements made to the data collection system and its integration into the COBATEST network. After 2013 a unique identifier was assigned to each client, allowing detection of repeat testers, and at the same time ensuring the anonymity of people tested. This could have led to an under-estimation of the number of tests in the period 2014-2018, especially in the MSM group as MSM are more often repeat testers. Secondly, the number of HIV positive cases from1995 to 2007 refers to HIV confirmed cases only, as the test offered was the conventional laboratory test. Since the introduction of HIV POCT, the number of HIV positive cases refers to reactive cases as in some cases, the information related to referral and confirmation of the diagnosis is not complete. So the number of HIV reactive cases in the period 1995-2006 could be higher. Lastly, regarding the number of cases detected in the DEVO network as a percentage of the total number of Catalonia HIV registry cases -as it was assumed that all reactive cases were linked to care and therefore were added to the Catalonia HIV registry, the contribution could have been overestimated.

Conclusion
In conclusion, our study with a monitoring series of almost 25 years reinforces the important role that community-based HIV POCT has on the improvement of early HIV diagnoses in key populations, and highlights the importance of monitoring these data and including them in a regional or national HIV surveillance system. To ensure sustainability of the community testing services, key stakeholders must commit to including CBVCT services in the design and plan for strategies to achieve the 90-90-90 objectives.