3.1. Background
Recent attention has focused on the question of how quickly ART should be started once HIV diagnosis is confirmed. In the early years of the HIV response, limited resources and concerns about suboptimal adherence led to a cautious approach in which people living with HIV underwent multiple counselling sessions that could last several weeks or months before starting ART (61). During this pre-ART period, substantial mortality and loss to follow-up were observed, especially among people with advanced HIV disease (62,63). This prompted research to focus on whether approaches to support rapid ART initiation, including initiating ART on the same day HIV infection is diagnosed or eligibility is determined, could reduce loss to care before ART starts and improve clinical outcomes.
The results of several recent randomized trials have indicated that rapid ART initiation, including same-day start, can improve programme outcomes, especially by reducing loss to care in the pre-ART period (64,65). Some evidence from programme settings, however, indicates that rapid initiation could lead to increased loss to follow-up after initiating ART because of insufficient time to accept and disclose HIV status and to prepare for lifelong treatment (66). This more rapid approach to ART initiation is particularly relevant to people with advanced HIV disease in consideration of the specific clinical benefits (reduced risk of mortality and morbidity) and potential harm (elevated risk of immune reconstitution inflammatory syndrome).
3.3. Clinical considerations when implementing rapid ART initiation or same-day initiation
People presenting for the first time or those returning to care should undergo history and clinical examination to evaluate for significant opportunistic infections (such as signs and symptoms of TB and signs and symptoms suggesting meningitis) before rapid ART initiation is offered. Although no longer a requirement for ART initiation, baseline CD4 cell count testing should be performed to determine whether the patient has advanced HIV disease.
People who have no clinical signs and symptoms of TB or other opportunistic infections and whose cryptococcal antigen test is negative may initiate ART the same day in combination with their package of prophylaxis outlined in Chapter 2. For people with CD4 cell count <100 cells/mm3 in settings where cryptococcal antigen testing result is not available on the same day, consideration could be given to starting fluconazole prophylaxis and discontinuing if a cryptococcal antigen screening result is subsequently found to be negative.
Previous WHO recommendations on the timing of ART initiation in the presence of TB and cryptococcal disease should be considered. These are as follows:
Timing of ART for people with TB
Routine TB symptom screening for people with HIV, using an algorithm containing fever, cough of any duration, weight loss and night sweats, will help to identify people who should either be expedited for TB diagnosis (if symptoms) or given preventive TB therapy (if no symptoms). Where feasible, suspected TB should be confirmed through laboratory testing (Xpert® MTB/RIF as the first test and LF-LAM in urine). ART should be briefly delayed while investigating for TB among people with TB symptoms.
TB treatment should be initiated first, followed by ART as soon as possible within the first eight weeks of treatment (strong recommendation, high-quality evidence).
TB patients
4 living with HIV who have severe immunosuppression (such as CD4 cell counts <50 cells/mm
3) should receive ART within the first two weeks of initiating TB treatment.
Caution is needed for people living with HIV with TB meningitis, since immediate ART is associated with more severe adverse events than initiating ART two months after the start of TB treatment.
Any child with active TB disease should start ART as soon as possible and within eight weeks after initiating TB treatment (other than TB meningitis
5), regardless of CD4 cell count and clinical stage
(strong recommendation, low-quality evidence).
Timing of ART for people with cryptococcal meningitis
Immediate ART initiation is contraindicated among people living with HIV with cryptococcal meningitis because of the risk of life-threatening immune reconstitution inflammatory syndrome (conditional recommendation, low-quality evidence).6
ART initiation should be deferred until there is evidence of a sustained clinical response to antifungal therapy and after four weeks of induction and consolidation treatment with amphotericin B–containing regimens combined with flucytosine or fluconazole or after 4–6 weeks of treatment with a high-dose fluconazole induction and consolidation regimen (conditional recommendation, low-quality evidence).
For people with signs and symptoms of meningitis, ART should be delayed pending the results of lumbar puncture.
There is no prospective evidence to support decisions about when to start ART among asymptomatic people with cryptococcal antigenaemia after initiation of pre-emptive antifungal therapy. Guidelines from the Southern African HIV Clinicians' Society (
67) recommend starting ART two weeks after starting fluconazole, and consideration is being given to starting ART immediately if lumbar puncture excludes cryptococcal meningitis in people who test positive for blood cryptococcal antigen.
Rationale and supporting evidence
The widespread adoption of the recommendation to treat all people living with HIV with ART regardless of clinical stage or degree of immunosuppression has identified a need to re-examine how ART is initiated. To date, standard preparation for ART initiation has included several pre-ART counselling sessions that must be completed before initiation; this period of preparation may last several weeks or months, and although it enables screening for and treatment of opportunistic infections, it also places an additional burden on both people and providers, resulting in avoidable loss to care (68–70).
To assess the impact of a rapid initiation schedule, including the offer of same-day initiation, a systematic review was performed examining data from three randomized controlled trials (64,65,71), 11 observational studies (68,72–81) and five qualitative studies (66,82–85). Among the randomized controlled trials, two individual trials were carried out among non-pregnant adults; in the single cluster-randomized trial, about one quarter of the women were pregnant (64). Combined interventions including point-of-care CD4 cell count and a revised counselling intervention were provided as part of the intervention arm in two trials (64,65). Two of the trials offered ART initiation on the same day that HIV was diagnosed; the other trial offered ART initiation on the same day of the first HIV-related clinic visit following a positive HIV diagnosis. The observational studies provided additional data on pregnant women (72,73,77–79) and people with acute HIV infection (74–76).
Across the randomized trials with rapid initiation, the likelihood of starting ART within 90 days of eligibility (pooled RR 1.3, 95% CI 1.23–1.39) and within 12 months of eligibility (RR 1.09, 95% CI 1.05–1.13) was increased. Retention in care at 12 months (RR 1.12, 95% CI 0.99–1.28), viral suppression at 12 months (RR 1.18, 95% CI 1.08–1.29) and mortality (RR 0.47, 95% CI 0.24–0.93) were influenced positively. In the observational studies, offering rapid ART initiation resulted in a greater likelihood of having started ART within three months (RR 1.53, 95% CI 1.11–2.10). However, in the observational studies, there was no evidence that offering rapid ART resulted in a greater likelihood of remaining in care (RR 0.97, 95% CI 0.79–1.18), and the risk of being lost to follow-up after ART initiation tended to be increased (pooled RR: 1.85, 95% CI 0.96–3.55).
Comparing benefits and harm
Linking people testing positive for HIV to ART services is programmatically challenging (62,86). The offer of rapid initiation, including same-day ART, increases the number of people starting ART, reduces mortality, and may further reduce both mother-to-child transmission and transmission to HIV-negative partners. However, possible harm identified includes the potential for missing clinical conditions requiring management before ART, the risk of immune reconstitution inflammatory syndrome among severely immunosuppressed people and the potential for people to feel coerced to start when they are not ready psychologically. Where health-care workers feel pressured to meet targets for ART initiation, this may lead to undue pressure on people to start ART as soon as possible. Special consideration should also be given to women in some settings where they may not be in a position to take the decision to start lifelong therapy independently.
Cost and cost–effectiveness
Three studies within the systematic review reported the cost and cost-effectiveness of rapid ART initiation. In a study among pregnant women in South Africa, rapid initiation was found to be very cost-effective compared with standard services: US$ 1160 per quality-adjusted life year saved (87). A second study, also from South Africa, found same-day treatment initiation to be both more effective and more expensive but noted that the use of the current point-of-care CD4 cell count technology predominantly drove the increased costs (88). In a low-prevalence study setting in China, the unit cost for an additional person receiving ART under the simplified testing and treatment approach was US$ 83.80, declining to US$ 9.69 in the second year; this was reported as an effective and sustainable intervention for the setting (81).
Equity and acceptability
Rapid initiation may improve the equity and accessibility of ART for people who may otherwise be lost to follow-up during ART preparation sessions (65). Nevertheless, in settings with limited ART access, setting priorities for initiating ART should still be based on clinical triage, ensuring that those with the most advanced HIV disease access care. Guiding principles highlighted within a community-led consultation on rapid initiation called for ensuring accurate information to be provided and the decision to start ART to be a collaborative process between the health-care worker and the person living with HIV (89).
Barriers and concerns regarding the acceptability of same-day ART initiation reported by people living with HIV included insufficient time to process information for pregnant women (82), limited time to disclose HIV status that could potentially result in stigma and conflict, including domestic violence (66), a need among pregnant women to seek approval from their partners before starting ART (83) and uncertainty about the HIV test result and the need for confirmatory testing (83). Assessment of these factors should be carefully monitored as the “treat all” policy is scaled up.
The acceptability of rapid ART initiation in specific populations must also be considered. To date, the evidence on rapid initiation has been derived only from adults. Preparing children and their caregivers to initiate ART, especially when syrups are prescribed, may require additional support. Adolescents and young adults should be included in the decision to start ART, since exclusion from this process may lead to adherence problems. Key populations should be offered rapid initiation but may have additional medical and psychosocial issues as well as adherence support issues before starting. For people who inject drugs, needle and syringe programmes and opioid substitution therapy should be given priority, but opioid substitution therapy should not be a prerequisite for initiating or maintaining ART for people who use opioids. Rapid initiation should also be offered to people returning to care. People should be assessed in a non-judgemental manner, discussing the reasons why they disengaged from care and stopped ART and giving the opportunity to restart ART rapidly, including on the same day if appropriate.
Feasibility
Health-care workers and programme managers reported rapid or same-day initiation as being feasible across all populations, despite some specific challenges in key populations. Adherence and time constraints within the clinic to perform all procedures required, including HIV re-testing before ART initiation, were the main issues raised regarding feasibility. One trial reported similar barriers such as the need for clinical confirmation of TB, having a WHO clinical stage 3 or 4 condition requiring treatment, insufficient time to complete all steps on the same day and individual preferences (65).
Access to baseline investigations may affect the feasibility of same-day ART initiation. Baseline CD4 cell count should be evaluated and the package of interventions for people with advanced HIV disease should be provided on the same day. However, the inability to perform such tests or the potential for results to be delayed should not delay the initiation of ART, except where this is clinically indicated (clinical suspicion of TB or cryptococcal meningitis, in which ART should be delayed to first investigate and decide on treating these conditions). TB preventive therapy, cotrimoxazole prophylaxis and, where indicated, fluconazole pre-emptive therapy and ART should be started at the same time.
Implementation considerations
WHO recommends that HIV positivity should be confirmed with a second specimen and, where circumstances allow, by a second operator using the same testing strategy and algorithm. This approach should be maintained in settings in which rapid ART initiation is being implemented to minimize the risk of misdiagnosis.
The 2016 WHO consolidated ARV guidelines (29) made recommendations on where and by whom ART initiation may be performed, including decentralizing ART initiation to peripheral health centres and task sharing to trained non-physician health-care providers, including midwives and nurses. For implementing rapid ART initiation, including same-day start, decentralization and task sharing should still be encouraged, ensuring adequate training in revised counselling schedules. ART initiation for people with advanced HIV disease may be feasible at decentralized sites and will be enabled through access to point-of-care diagnostics. Non-physician health-care providers with adequate training may initiate ART rapidly but should have clear referral criteria to ensure that people access appropriate investigation and specialist management.
Rather than giving HIV and ART education over several sessions before ART initiation, the timing of counselling should be adapted. Where initiation is to be performed on the same day, priority should be given to how to develop an immediate adherence plan and how to recognize side-effects. There is evidence that establishing good adherence in the initial period following ART initiation is important for long-term treatment success (90). Further counselling to support treatment literacy, including the need for lifelong optimal adherence, how ART is monitored, and options for future differentiation of care should be covered in subsequent counselling sessions during the first months on ART. For those clients who are not ready for same-day start, the content of the ART preparation should be given over the following seven days, while considering the ability of the person to travel to the clinic. Where community health workers or expert clients are trained in ART preparation counselling, this may also be performed at the community level (68). Flexibility in how these sessions are delivered requires adapting them for the people who need more preparation time to start ART. Checklists to support clinicians in assessing people's clinical and psychological readiness should be developed, and the impact on workload should be assessed further.
The availability of point-of-care diagnostics for CD4 cell count, TB diagnosis and cryptococcal antigen lateral flow assays (whole blood, plasma or serum) may also support the programmatic implementation of the package of care for people with advanced HIV disease, enabling such people to access appropriate prophylaxis or pre-emptive treatment when this is indicated and to allow rapid investigation for TB before ART initiation.
Finally, initiation processes need to be adapted to recognize the increasing number of people that will re-enter care after a period of treatment interruption. For example, to overcome stigma that may be associated with return to care, Welcome Back clinics have been established in South Africa to encourage those who have stopped ART to return and be provided with support to continue therapy.
3.4. Considerations for children living with HIV who are hospitalized or severely ill
Two recent trials have suggested that very rapid initiation of ART may not be appropriate for sick children. In the first study from South Africa, young children (median age 23 months) with severe acute malnutrition were randomized to receive ART within 14 days of admission or delay it until nutritional recovery. The results suggested that delayed ART improved immune recovery, viral suppression and anthropometric measures, but the treatment arms did not differ with respect to mortality (91). These results contrast with the results reported by a retrospective, non-randomized study in Malawi, which found that initiating ART within 21 days of outpatient therapeutic feeding was associated with improved outcomes among children living with HIV with uncomplicated malnutrition in Malawi (92). In the second study in Kenya, hospitalized children living with HIV (median age 23 months) were randomized to start ART within 48 hours versus 7–14 days. Although treatment arms did not differ with respect to mortality, the authors concluded that rapid treatment (whether immediate or within 14 days) is safe, and prompt initiation of ART is essential to reduce the very high mortality observed overall, with 21% of children dying during six-month follow-up (93). Overall, although ART initiation remains a priority, especially for children younger than five years and children who present with symptoms, timely provision of appropriate care for clinical conditions requiring acute management is the first priority.
