HIV remains a critical public health challenge [1] [2] , especially in sub-Saharan Africa region, which accounts for nearly two-thirds of the global burden [3] . Since 1981, an estimated 85.6 million people have been infected and 40.4 million have died, ranking HIV among the deadliest pandemics in human history [4] . Mozambique has one of the highest HIV prevalence rates in the region, with 12.5% among adults and higher levels in urban areas, such as Maputo [5] . Over the past decade, substantial progress in ART scale-up has been made, followed by national policy reforms to improve access, retention, and survival of people living with HIV (PLHIV) [6] [7] . The 2015 introduction of the World Health Organization (WHO) Test-and-treat strategy (TTS) marked a paradigm shift in HIV care.

Evidence from observational studies and randomized trials across resource-limited settings consistently shows that early ART initiation improves clinical outcomes, reduces HIV incidence and decreases HIV-related morbidity and mortality [8] - [10] . However, social epidemiology underscores the importance of contextual factors in shaping the effectiveness of public health interventions such as TTS [11] - [13] . Despite notable progress following TTS implementation, major challenges persist, with adherence to ART, retention in care, and survival outcomes remaining suboptimal in many settings [14] - [16] .

Studies from across sub-Saharan African (SSA) countries indicate that although the TTS has improved the overall survival, a substantial proportion of patients still experience early mortality, particularly within the first year of treatment [17] - [19] . Poor outcomes have been consistently associated with advanced clinical stage at treatment initiation, late presentation to care, male sex, and comorbidities such as tuberculosis [20] [21] . In Mozambique, this may also be the case; however, evidence remains limited, and little is known about survival among patients initiating ART under TTS, highlighting the opportunity to evaluate treatment outcomes in routine programmatic settings. Understanding survival outcomes and their predictors is essential for optimizing HIV programs, guiding resource allocation, and informing interventions to reduce preventable deaths. This study assessed survival among HIV-positive patients initiating ART under TTS and identified factors associated to survival, in Maputo.

The study was conducted between 2013-2020, during which a total of 16,968 ART patients were included, established in two cohorts: 9036 (53.3%) enrolled before TTS introduction (BTT) and 7932 (46.7%) after TTS introduction (ATT). Baseline characteristics were generally comparable between cohorts, except for sex distribution, with fewer females in the ATT cohort (41%) than in the BTT cohort (65%), likely due to temporal enrolment patterns. The distribution of other characteristics was generally comparable between both cohorts ( Table 1 ). Kaplan–Meier curves indicated higher survival among patients in the ATT cohort compared with those in the BTT cohort ( Figure 3 ).

BTT (Jan/2013-June/2016); **ATT (Jan/2017-Dec/2019); ***First CD4 count after diagnosis; ^#Community ART groups; n = simple size.

To our knowledge, this is the first cohort study in Mozambique to estimate the effect of the TTS on survival among patients receiving ART in public HFs. Using survival analysis, we compared survival probabilities and mortality rates among patients initiating ART before and after the implementation of TTS in Maputo.

HR = crude hazard ratio; aHR = adjusted hazard ratio; ART = antiretroviral therapy; ^bFirst CD4 count after diagnosis.

In this study, patients exposed to TTS demonstrated higher median survival time and greater survival probabilities at 6, 12, 18, 24, and 36 months compared with those not exposed. The mortality rate was also lower in the post-TTS cohort (24.5 per 1000 person-years) than in the pre-TTS cohort (28.1 per 1000 person-years), corresponding to a 12.8% reduction following implementation. These findings are consistent with evidence from other settings. For example, a study in southern Ethiopia reported higher survival probabilities and lower mortality (2.4 per 100 person-years) among TTS-exposed patients, compared to 3.8 per 100 person-years among non-exposed patients [14] . Similarly, in Addis Ababa, mortality rates were 81.8 versus 92.4 per 1000 person-years among TTS-exposed and non-exposed patients, respectively [26] . Comparable trends have been observed in Eastern Uganda [27] and in Chengdu, China, where the study focused on men who have sex with men (MSM) [28] . Furthermore, a cluster-randomized community trial in KwaZulu-Natal, South Africa, demonstrated a 31.6% reduction in mortality, from 17.1 to 11.7 per 1000 person-years between the control and intervention groups [29] .

Analysis of associated factors indicated that survival decreased with increasing age in both cohorts, with older patients experiencing a higher hazard of death compared to those aged 15 - 24 years. Notably, age-related differences in mortality narrowed following the introduction of TTS. These findings have important clinical and programmatic implications, underscoring the need for tailored interventions for older patients who remain at elevated risk. Similar age-related survival patterns have been documented in Ethiopia [30] and Brazil [31] . Disparities by region of residence were also evident, with higher mortality among patients living in suburban areas, consistent with observations from Gurage, Ethiopia [14] .

HIV/TB co-infection was another significant predictor of survival, with co-infected patients experiencing a higher hazard of death than those without TB; nevertheless, the risk decreased following TTS implementation. These findings align with evidence from a systematic review and meta-analysis, as well as studies from southern Zambia and northwest Ethiopia [32] - [34] . Membership in CAGs was also associated with improved survival, highlighting the protective role of differentiated service delivery models and corroborating findings from Mozambique, Nigeria, and Uganda [35] - [37] .

A key contribution of this study is the estimation of the adjusted PAF for the exposure. The PAF of 7.3% indicates that a modest but meaningful proportion of deaths in the study population were attributable to the absence of TTS. Although the impact on mortality was relatively smaller, these findings highlight the survival benefits of TTS among ART patients in Maputo. Overall, the findings suggest that widespread TTS implementation substantially improved survival in this study population. To date, no other cohort study in Mozambique has reported comparable findings. From a public health perspective, these results underscore the potential of TTS to improve ART outcomes and offer valuable evidence to guide policymakers and HIV program implementers in Mozambique.

This study is significant in three main areas: 1) it is among the first to evaluate the impact of TTS on survival among ART patients in Mozambique, providing evidence to inform the National HIV & AIDS Control Program; 2) it confirmed survival predictors identified in other sub-Saharan African contexts, reinforcing their relevance to the Mozambican setting; and 3) it highlighted the importance of monitoring undocumented mortality and patient self-transfer, which continue to challenge ART program success.

The study has several limitations. First, it relied on secondary data from public health service databases, which are often incomplete and may not fully represent individuals from higher socioeconomic backgrounds. Second, undocumented mortality and self-transfer could have led to overestimation of hazard ratios. Finally, the relatively short follow-up period for each cohort may have limited the ability to capture longer-term effects of TTS implementation

Test-and-treat strategy improved survival and reduced mortality among ART patients in Maputo, highlighting its relevance for Mozambique and other resource-limited, HIV high-burden settings. Future research should examine its effect on specific subgroups, undocumented mortality and self-transfer.

The study was fully designed, conducted, and reported by the primary author, with the co-author providing substantial contributions to the review process as required.

We thank the Maputo City Health Services (MCHS) for granting access to ePTS data and the health facility staff for supporting in data extraction from Open MRS.

No external financial support was received for this study.