You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

HIV treatment cascade

HPTN 071 attempts universal home-based HIV testing in sub-Saharan Africa: scaling the mountainous challenge of UNAIDS 90-90-90 target

9 Jun, 17 | by Leslie Goode, Blogmaster

Estimates of 96% for the preventative efficacy of ART against HIV transmission, reported in 2011 by Myron & Cohen (M&C), appeared at last to place long-term containment of the epidemic in our hands.  In the wake of this, UNAIDS: 90-90-90 proposed ambitious targets: 90% of those living with HIV to know their status; 90% of known HIV+ individuals to undergo ART initiation; 90% of ART initiated to achieve viral suppression.  Yet – for all the promise of known ART efficacy – the question remains how easy it will be to realize these targets in practice, especially in those settings (e.g. sub-Saharan Africa) where the impact of the epidemic is at its most severe.

Perhaps less easily than one might imagine, suggest Hayes & Fidler (H&F) – a first (one year in) report of results from HPTN-071.  This is a large cluster-randomized study of home-based HIV testing and treatment in 21 Zambian and S. African communities.  The results derive specifically from four of the seven ‘intervention’ communities situated in Zambia.  Of the c.100,000 consenting to the intervention, and visited by pairs of Community HIV Care Providers (CHiPs), 77% of men and 85% of women ended up knowing their status.  (The rest were presumably not reached, or refused testing.)  The estimated proportion of known HIV positive individuals who then went on to initiate ART was 42% within six months and 53% within 12 months.  Over the first year, the estimated overall percentage of HIV+ adults in the four communities who were on ART rose from 44% to 61%.  Obviously, these results reflect only the first year of the intervention, and CHiPs will revisit participants.  At this stage no estimate of the third UNAIDS target (viral suppression) is possible.  Nevertheless, the results reported by this paper give us a sense of the very considerable challenge that the achievement of the UNAIDS target represents in practice.  Among the difficulties, H&F refer specifically to the difficulty of accessing male participants at their homes, and the very slow rates of initiation to ART.

There are various ways of delivering testing.  Home-based testing has been trialled in populations that would otherwise be very difficult to access – e.g. in indigenous Amazonia (Ribeiro & Benzaken (STIs) (R&B); Benzaken & Peeling (STIs) (B&P).  In the case of HPTN-071, its attraction seems to be the possibility of most easily achieving quasi universal coverage.  At 90-90-90, achieved rates of viral suppression would be 73%; at 80-80-80, they would only be 43% – and the preventive efficacy of ART would be correspondingly reduced.  Yet even the latter target seems ambitious.  So far as the testing is concerned, however, the 90% is within reach – at least in Zambia.  The real challenge will be second (and probably also the third) stage of the cascade.  Other studies indicate that HPTN-071 is not unique in seeing large fall out at this point (see Schwartz & Baral (STIs)).  R&B and B&P point to further problem likely to arise at the stage of rolling out this kind of preventative ART intervention – namely that of maintaining the quality of POCT testing (Benzaken & Peeling (STIs)).


The PrEP ‘care continuum/cascade’: how would it look?

8 Mar, 17 | by Leslie Goode, Blogmaster

We take for granted the value of the care continuum (or ‘cascade’), now increasingly seen as the key measure of health system response to HIV (Cassell (STIs editorial)).   The application of this model to HIV has provided a benchmark for evaluation in contexts as diverse as Moscow (Wirtz & Beyrer (STIs)), South Africa (Schwartz & Baral (STIs)) or the Netherlands (van Veen & van der Sande (STIs)).   But could the same model also offer a means of evaluation in the case of other complex sexual health interventions such as PrEP (Pre-Exposure Prophylaxis)?

An on-line soon-to-be-published paper by Nunn & Chan (N&C), building on an earlier attempt by Kelley & Rosenberg (K&R), does precisely this.  An important difference from the earlier paper seems to be the more concrete definition of a larger number of steps (nine as against five) – especially in the central area of ‘uptake’ and engagement in care.  Here K&R define three stages: ‘need for awareness of PrEP and willingness to use it’, ‘need for good access to healthcare’, and ‘need for a prescription for PrEP.  N&C replace these with a more concrete conceptualization of the process in five stages involving: an occasion where PrEP access is facilitated (4); an appointment arising from that occasion where the assessment is performed (5); the prescription of PrEP, where indicated (6); the actual initiation of PrEP (i.e. when the client starts taking the pills) (7).  Also important is N&C’s substitution of two final steps – adherence (8)) then retention (9) for K&R’s single final step of ‘adherence’.  N&C point out that, whereas, with ART, ‘adherence’ is once-and-for-all and secures the ultimate goal of viral suppression, in the case of PrEP, we can envisage multiple trajectories depending on whether PrEP continues to be indicated (e.g. the client may no longer be exposed to risk).  Finally, K&R’s first step – ‘identifying at risk MSM’ – gives way to three: identifying at risk individuals (1), enhancing HIV awareness (2), enhancing PrEP awareness (3).

Is this nine-stage definition of a PrEP cascade overly “complex” (EECAAC2018)?

Answering such a question requires us to reflect on the function that the ‘cascade model’ is called upon to perform.  If the model divides up the total course of an intervention into a series of staged tasks, this is presumably because the health benefit depends on the completion of the whole intervention, yet the accomplishment of each step is necessary to the achievement of subsequent ones.  The idea of the cascade can provide a fair way of evaluating the progress of an intervention before its potential health benefits have been delivered – and can also identify the precise points at which the intervention is failing (i.e. where clients become ‘disengaged’).

It follows that each step should correspond to a potential outcome that is not inferable from previous or later outcomes but is worthy of independent evaluation.  If everyone who accesses PrEP (4) also attends an appointment at which suitability of PrEP is discussed (5), or everyone who adheres to PrEP (8) is also retained in PrEP (9), then steps (4) and (5), or steps (8) and (9), can be merged.  This is not stated in so many words by the authors of the model.  However, I would assume that it must lie at the basis of their thinking.

A case for immediate ART initiation in all HIV diagnosed, regardless of CD4+ count?

28 Aug, 15 | by Leslie Goode, Blogmaster

The year 2013 saw the WHO recommended threshold for treatment of HIV+ patients with ART rise from a count of 350 CD4+ per mm3 to one of 500 CD4+ per mm3.. The threshold had been fixed as early as three years before (2010) at 350 CD4+ per mm3 (WHO ART recommendations). These recommendations have doubtless been made with a view both to improving individual outcomes and preventing transmission. However, their implementation poses, and will continue to pose, a serious challenge, especially in resource-poor settings. Writing in 2010, Hamilton and Crowley (STIs) estimated that setting the threshold for ART initiation at 250 CD4+ would by 2012 increase the need for treatment by a median of 15%, whereas setting at 350 CD4+ would increase it by 42% and 500 CD4+ by 84%. Contributors like Hallett & Garnett (STIs) (Zimbabwe) and Zwahlen (STIs) have sought to develop projections for individual countries.

It may not be so long before the modellers have, once again, to revise their calculations. Results of the Strategic Timing of Antiretroviral Therapy (START) trial, a recent multi-continental RCT, make a case for not deferring ART initiation, regardless of the patient’s CD4+ count. A total of 4,865 patients in 35 countries were randomized to an immediate and a deferred initiation group, and followed over a three-year period. The trial compared primary end-points, including death and serious AIDS-related, and non-AIDS-related, health events, over a three-year period. Among the commonest AIDS-related events were tuberculosis, Kaposi’s sarcoma and malignant lymphoma. Non-AIDS related events included cardio-vascular disease and non-AIDS related cancers. Hazard ratios for serious AIDS-related and non-AIDS related, events were 0.28 and 0.61, respectively. There have been concerns about the negative health impact of ART especially on cardio-vascular disease. But no safety concerns in the immediate-initiation group were identified. However, as the authors recognize, the study was underpowered for investigating the potential negative effects of early ART initiation on individual non-AIDS related events, because of the relatively limited number of events and early termination of the deferred therapy strategy.

In principle, the benefits of raising the threshold for linkage to care are clear. In due course, we may expect to see yet another upward adjustment in the WHO threshold. However, recent studies point to inadequate availability of care, late diagnosis (even by current standards), and poor retention in care as significant factors in suboptimal health outcomes (Mubezi & Shuha (STIs); Hussey (STIs). So what such a change in guidelines would deliver in real terms is hard to evaluate, and no doubt depends on the context of implementation. Yet, even in the relatively more resource rich settings (US), some recent research has argued for the reallocation of resources from linkage to retention in HIV care, in order to optimize utilization of scarce resources (Retention in Care (STI/blogs); Sherer (STIs). This recommendation would seem hard to square with the prioritization of ever lower thresholds for linkage to care.

At all events, this study demonstrates the benefits to the individual of prompt linkage to care whatever the stage in the progression of his/her infection.

Retention in care rather than diagnosis may prove the ultimate challenge for US HIV response

25 Mar, 15 | by Leslie Goode, Blogmaster

The real challenge which the US HIV/AIDS epidemic poses for the US public health services is not simply to achieve higher levels of diagnosis – but, far more than that, to improve linkage to, and retention in, care.  This claim is hardly controversial. But it is thrown into stark relief in a recent study by Skarbinski & Mermin, which estimates the number of HIV transmissions attributable to non-retention in care for 2009.

The authors employ the notion of a five-phase “care continuum”.  Using population data from the National HIV Surveillance System and medical data from the National HIV Behavioral Surveillance System and the Medical Monitoring Project, they estimate the number of HIV transmissions occurring at each phase.  The phases in the continuum are: (1) infected, but undiagnosed; (2) diagnosed, but not retained in care (attending at least one visit to a medical care provider Jan. – April 2009); (3) diagnosed, retained in care, but not given ART; (4) diagnosed, retained in care, prescribed ART, but not virally suppressed; (5) virally suppressed.

The reduction in attributable transmissions achieved for those diagnosed but not retained in care (phase 2), as compared with those who remain undiagnosed (phase 1), is 19%.  (It is probably due to a decrease in HIV-discordant unprotected sex).  But the reduction achieved for those who achieve viral suppression (phase 5), as compared with those who remain undiagnosed, is 94%.  In estimating the epidemiological impact of these reductions, we need to factor in the percentage of the infected population at each phase.  The large proportion (45.2%) of the HIV infected who are diagnosed but not retained (phase 2) explains the very high proportion of total transmissions (61.3%) attributable to this phase.  By comparison, only 30.2% are attributable to the undiagnosed (phase 1), and 2.5% to the virally suppressed (phase 5).  The low epidemiological impact of those at phases 3 and 4 is due to the relatively low proportion of those infected who remain in these phases.

The message, then, is that achieving greater success in retaining the HIV diagnosed in care may prove the key to combating the epidemic at population level.  Of course, diagnosis remains the indispensable first step.  But the potential gains of diagnosis will be only very partially experienced, so long as such a large proportion of those diagnosed are not retained in care.  Of course, improving retention in care may constitute a somewhat different – and perhaps more difficult – challenge for the US health services from diagnosis.  The specific problems of the US health system in this regard are discussed by Sherer (STI), and the characteristics of individuals “lost to follow up” by Haddow & Mercey (STI) and Lee & Gazzard (STI).  Local attempts to address these problems through a more “wrap-around” approach to health care in the US are described in my blog Bocour & Less (STI/blog) (see Bocour &  Less).  There has also been interest in the computer assisted self-interviewing in order to engage those lost to care (Dombrowski & Golden (STI)).

Sexually Transmitted Infections blog

Sexually Transmitted Infections

Discussion and suggestion space for readers of STIs.
Visit site

Creative Comms logo

Latest from Sexually Transmitted Infections

Latest from Sexually Transmitted Infections