This study explored the resumption of HIV services for AYAs in the period after the end of the state of emergency following the COVID-19 pandemic in Mozambique. Both the number of outpatient visits and HIV testing showed a downward trend compared with the steady increase during the COVID-19 pandemic returning to values comparable with the pre-pandemic period. Throughout the study period, the HIV positivity ratio decreased, reaching 1.7% in the post-pandemic period. Similarly, the non-adherence ratio decreased across the three periods, reaching the lowest value in the post-pandemic period. The recall ratio increased across the three periods, while the return ratio did not show any significant trend despite a decrease during the pandemic period. There was no evidence that the location influenced the indirect impact of the pandemic on the outcome measures.

As already highlighted in the literature, the COVID-19 pandemic had a negative indirect impact on the provision of HIV services for PLHIV in many different countries [20]. In SSA, the major impact occurred between the second and third quarter of 2020 with a reduction of 33.5% and 30.8% in number of HIV test and new ART initiations, respectively, in adolescents aged 10–19 years [21]. On the other hand, an increase from 78.5% to 80.5% in viral load suppression was observed between second and third 2020 quarters [21].

According to our results, both the number of HIV tests and the ART pick up, a proxy measure of adherence, showed an ameliorating trend during the two years of the pandemic, although the HIV tests experienced an initial decline at the beginning of the pandemic. This is in line with findings in other studies where ART services were prioritised and generally maintained even during the early stages of the pandemic while HIV testing was more heavily affected [5, 22]. The constant ameliorating trend found in the present study during the pandemic period for the number of outpatient visits and tests, and for non-adherence and recall ratio could be also due to the fact that the most strict measures were taken during the first wave of the COVID-19 pandemic in Mozambique, similarly to other countries in SSA, while during the second wave, despite its greater severity, there was not the same public health response [23].

In the Fourth round of the global pulse survey on continuity of essential health services during the COVID-19 pandemic conducted by the WHO between November 2022 and January 2023, a general recovery of services was reported in this quarter [24]. In particular, the percentage of disruption of services compared to the pre-pandemic period decreased from 48% in the first quarter of 2021 to 24% in the quarter analyzed for HIV testing and from 32 to 8% for the continuation of established ART [24]. Based on these data, not only there was evidence of a reduction in service disruption, but some countries also experienced an improvement in some HIV services. In particular, an increase over pre-pandemic levels in adherence to ART and in the number of HIV tests was reported by 15% and 9% of the countries surveyed, respectively [24]. Similarly, our findings showed a reduction in the non-adherence rate at the beginning of the post-pandemic period. Both the number of outpatients visits and the number of tests, however, showed a downward trend in the post-pandemic period compared to the pandemic one, although not different from the pre-pandemic period. This could be attributed to a rebound effect after the initial drop at the beginning of the pandemic. Nevertheless, this trend should be closely monitored to avoid losing the positive results obtained during the pandemic. The 2023 executive summary by the Joint United Nations Program on HIV/AIDS (UNAIDS) shows that a total of US$ 20.8 billion was allocated for HIV programs in low- and middle-income countries in 2022. This is 2.6% lower than the funds allocated in 2021 and much lower than the estimated US$ 29.3 billion needed [25]. The provision of adequate economic resources and funding for HIV programs is crucial to maintaining positive trends in the health services delivery. Indeed, although the COVID-19 pandemic has brought a partial increase in investments in the health sector, this must be maintained and adapted to be sustainable and to support health services where it is needed such as HIV services in SSA countries [26]. Maintaining the necessary funds after the COVID-19 pandemic in these services would have a triple beneficial effect by improving the health, social and economic gains of the countries in a virtuous circle [27]. Especially allocating more economic resources to community-based programs is believed to be essential to achieve the 95-95-95 goals for HIV [27].

Among the HIV services most affected by the COVID-19 pandemic along with the number of HIV tests were community programs, such as community support groups, patient tracing activities and outreach activities [28]. The data from our study showed how re-engagement activities were heavily affected by the pandemic, with a decrease compared to the pre-pandemic period, and even at the beginning of the post-pandemic period, the recall ratio made by community health care workers was still at a lower level than in the pre-pandemic period.

The disruption of community activities and testing at the very beginning of the pandemic period may also explained the decreasing positivity rate observed during the study period. Community-based testing is a key pillar for reaching population groups at high risk of HIV. The systematic review on 16 PEPFAR-supported countries found a similar slight decrease from 1.9% before COVID-19 to 1.7% during COVID-19 [21]. The discontinuation of outreach and testing services in communities due to the pandemic suppressed many of the case-finding strategies for key populations. Among these there may be the AYAs who were most affected by the reduction in HIV testing and diagnoses in SSA [29].

In LMICs, the adaptation of HIV services mainly relayed on differentiated service delivery (DSD), in particular for ART services [30]. In Mozambique, managers and providers from the Ministry of Health acknowledged a positive effect of the COVID-19 pandemic, as it loosened eligibility criteria for enrolment in fast-track and 3-months ART dispensing [31]. DSD has the advantage of being a person-centred approach that simplifies and adapts HIV services throughout the cascade, reducing unnecessary burdens on the health system [31]. This model not only was effective in maintaining HIV services but also had the advantage of limiting the impact of COVID-19 infection in a fragile and at-risk population such as PLHIV [32]. Indeed, seroprevalence studies showed no significant differences in SARS-CoV-2 infection probability between the general population and PLHIV [33]. Throughout the pandemic period, HIV services were never interrupted and, as in other countries, Mozambique prioritised healthcare workers, as well as other at-risk groups, for vaccination for SARS-CoV-2 [34].

In the present study, no differences in HIV service delivery were found based on the HC being located in a central or peripheral area within an urban setting. This may be due to the fact that HCs located in rural areas were not included in the study, as these areas showed the greatest fragility and difficulties in HIV service delivery as they are also highly reliant on community-based activities [28, 35]

This study has some limitations. First, the pre-COVID period encompassed a time frame of less than one year and following cyclone Idai, so it may not represent a standard of normality reference for HIV services. Second, data on testing and outpatient visits were collected on an aggregate level, so we could not assess how many new people were reached by HIV services during the reporting period nor explore possible differences in access to care based on sex. Finally, data on HIV disease stage and viral load were not collected and analysed, which would also have enabled an assessment of the clinical status of PLHIVs over the three periods included in this study.