Can the “Seek, Test, Treat, and Retain” Strategy be Effective in France?

Abstract

To the Editors: INTRODUCTION On an individual level, linkage and retention to care are known to be critical for HIV-infected patients.1 On a population level, this will allow a decrease in the community viral load (VL) and consequently a decrease in HIV transmission.2–4 The Centers for Disease Control and Prevention estimated that the proportion of virologically suppressed among HIV-diagnosed patients in the United States was 35%,5 mainly due to issues such as access to care or adherence to care and treatment. Access to care and treatment is highly dependent on the health care system, and very different estimations have been made in a country with free access to treatment.6 In France, public hospitals provide free-of-charge care for all, and public insurance covers 100% of the antiretroviral therapy and other associated costs. The objective of this study was to estimate the proportion of virologically suppressed among the total number of diagnosed HIV-infected patients in 8 French regions between January 1, 2005, and December 31, 2010. PATIENTS AND METHODS First, we estimated the total number of patients with a first HIV-positive test during the study period in each region. We used epidemiological data from the national surveillance system Institut National de Veille Sanitaire to estimate the reported number of HIV infections diagnosed during the study period in Alsace, Franche-Comté, Martinique, Midi-Pyrénées, Nord-Pas-de-Calais, Pays de Loire, and Provence Côte d'Azur.7 Those reported numbers are considered to represent 72% of all HIV-tested cases8; this correction factor was used to estimate the total number of HIV-diagnosed patients. Second, we estimated the number of HIV-infected patients that had initiated care in the corresponding regions during the study period. Each geographical region has 1 highly specialized HIV reference center; other facilities providing care with possible reference to their regional specialized center. We used specialized HIV reference centers prospectively maintained databases9 and estimated the number of patients with a first HIV diagnosis during the study period that had at least 1 medical encounter in that HIV reference center. The proportions of patients in care in the specialized units among the total number of patients in care in a given region are made public yearly.10–15 Using these proportions, we estimated the number of patients that initiated care in the corresponding region during the study period. Third, we estimated the proportion of patients lost to follow-up. This was again estimated using specialized HIV reference center prospectively maintained databases and hypothesizing that this proportion was not different in other facilities that provide care in the region when compared with the specialized HIV reference center. Lost to follow-up was defined as patients who did not show up to clinic for at least 12 months, were not known to be under the care of another hospital during this period, and were not known to have died within 12 months of their last visit.16 Fourth, among those who initiated care and not lost to follow-up, we estimated the proportion of patients with an “undetectable” VL using the last available VL available in the databases and hypothesizing that this proportion was not different in other facilities that provide care in the region when compared with the specialized HIV reference center. Because it has been hypothesized that a VL below 1000 copies per milliliter may be sufficient to reduce transmission at least in heterosexual couples,17 we studied 2 thresholds for an undetectable VL: below 50 and below 1000 copies per milliliter (respectively, national recommendations for considering treatment as successful and risk reduction threshold). RESULTS Results are reported in Figure 1. In the 8 relevant regions, 4092 patients were reported to have a first HIV-positive test during the study period7; the total number of HIV-diagnosed patients was therefore estimated at 5683. In the HIV-specialized reference centers, 3077 patients diagnosed during the same period had at least 1 occurrence; from this, the estimated HIV-infected patients that had initiated care in the 8 relevant regions was estimated at 4512. The proportion of patients with regular follow-up in the HIV-specialized reference centers was estimated at 74% (2286/3077). The proportion of patients with VL below 1000 copies per milliliter and below 50 copies per milliliter among those regularly followed-up in the HIV-specialized reference centers was estimated at 92% (2114/2286) and 83% (1907/2286), respectively. We therefore estimated that 55% of the 5683 HIV-diagnosed patients in our regions may have VL low enough to reduce transmission.FIGURE 1: Patients diagnosed with HIV, linked to care, treated, and obtaining low VL in 8 regional centers in France.DISCUSSION In this study, we estimated the proportion of HIV-infected patients with an undetectable VL over the total number of HIV-diagnosed patients. We show that the public health objective of reducing transmission risk was reached for 55% of HIV-diagnosed patients in the study period in the studied regions. Our results are quite different from previous estimations from the United States,5,18 which reported that 35% of the patients with an HIV-positive diagnosis had an undetectable VL. This is most probably because of differences in health care systems. Of note, we do not consider the proportion of patients who are infected but not diagnosed because it is not possible to give estimates of this population at the regional level. We therefore do not explore HIV testing in France. But once tested, our data provide good information on the further steps needed for a treatment as prevention strategy. Although we show that the proportions of patients with an undetectable VL over the total number of HIV-diagnosed patients is higher than in the United States, we also show that 45% of diagnosed patients are either not linked to care or adherent to care and treatments. We estimated the number of HIV-infected patients that are diagnosed using epidemiological data from the national surveillance system. These data are anonymous for legal reasons. Therefore, we can not directly link HIV diagnosis and entry in care. It could be possible that we were analyzing very distinct populations, if at the extreme all people living in 1 region used to seek care elsewhere. However, most likely people seek care in the region in which they live. We also hypothesized that the proportion of patients lost to follow-up among those diagnosed and under care or the proportion of patients with an undetectable VL among those regularly followed-up were not different in other facilities that provide care in the region when compared with the regional specialized HIV reference center. Again, this is probably a realistic hypothesis. For example, in one of the considered geographical regions, it has been shown that the incidence of HIV-infected patients lost to follow-up was not different in other facilities when compared with the regional specialized HIV reference center.16 In a public health system with free access to care and treatment, the most challenging part of the “seek, test, treat, and retain” strategy remains the “seek and test” part.19 Nevertheless, patients also keep falling out of the system at some time after diagnosis. If we are to reduce HIV incidence in France, we do need innovative HIV testing strategies to reduce the number of patients ignoring their HIV status, but we do also need to enhance global care.