BackgroundLoss to follow-up (LTFU) among under-five children from HIV care profoundly affects the treatment outcomes of this vulnerable population. It is a major factor that negatively affects the benefits of antiretroviral therapy (ART). Current information about LTFU among HIV-positive under-five children on ART is essential for effective treatments. To far, nevertheless, limited research has been done in Ethiopia to address this issue. Thus, this study aimed to assess the incidence and predictors of LTFU among HIV-infected under-five children receiving ART in West Amhara Comprehensive Specialized Referral Hospitals.MethodsA multicenter institution-based retrospective follow-up study was conducted among 435 HIV-infected under-five children on ART selected by simple random sampling from January 1, 2010 to December 31, 2019, and data were collected between December 1, 2021, and January 31, 2022. A standardized data extraction tool adapted from the ART entry and follow-up forms was used. The event of interest for this study was LTFU, whereas the absence of LTFU was censored. Before being transferred to STATA version 14 for analysis, the data were entered into Epi-Data version 3.1. The Kaplan‒Meier curve was used to estimate an individual's survival-free probability at each specific point in time. The Cox proportional hazards model was used to identify predictors of LTFU.ResultsAmong the 420 records included in the final analysis, 30 (7.14%) of the individuals were LTFUs. The incidence rate of LTFU was 3.4 per 1000 person-months of observation (95% CI: 2.43–4.87). The survival probabilities of children after 12, 24, 36, and 48 months were 0.97, 0.92, 0.88, and 0.77, respectively. The independent predictors of LTFU were HIV infection in under-five children who lived in rural areas (AHR = 3.64; 95% CI: 1.41, 9.37), poor adherence to ART (AHR = 4.37; 95% CI: 1.59, 12.02), not receiving cotrimoxazole preventive therapy (AHR = 3.75; 95% CI: 1.39, 10.08), not receiving isoniazid prophylaxis (AHR = 3.4; 95% CI: 1.29, 9.01), and having a severe WHO clinical stage (AHR = 5.43; 95% CI: 1.38, 11.43).Conclusion and recommendationThe incidence of loss to follow-up was high, especially in the first two years after ART initiation. The risk of LTFU was greater for those who were rural residents, had poor adherence, lacked cotrimoxazole preventive therapy, not given isoniazid prophylaxis, and presented with WHO clinical stages III and IV. Therefore, clinicians should emphasize for cotrimoxazole preventive therapy and isoniazid prophylaxis, for those living in rural areas, who present with poor adherence and WHO clinical stages III and IV.