Background: Although the gold standard for diagnosing beta-lactam antibiotic (BLA) allergy is the drug provocation test (DPT), there is no standardized protocol for children. Objective: We aimed to evaluate the clinical features and DPT results of children with a history of low-risk non-immediate reactions (NIR) to BLA who underwent initial direct single therapeutic dose challenge with a 5-day prolonged DPT. Methods: We retrospectively evaluated children ages 0–18 years with a history of low-risk NIRs to BLAs. On the first day of provocation, a single-dose DPT protocol without any skin test was administered at the clinic. The therapeutic dose was adjusted to not exceed the maximum single-unit dose (MSUD) for age and weight. The DPT protocol was administered with 100% of MSUD. To identify children with delayed reactions, the parents or caregivers were told to continue giving the medication at home for 5 days. Results: One hundred and nine children were included in this study. The median (interquartile range) age of the children was 62.5 months (26.5–94 months). Of the suspected drugs, the main culprit drug was amoxicillin-clavulanic acid for 89 children (81.7%). The most common clinical manifestation was maculopapular exanthema, which occurred in 85 children (78%), and 8 (7.3%) had a positive DPT result. Three children (2.8%) developed a reaction after the first DPT dose. The remaining children continued to use the suspected BLA at home. Five children (4.7%) developed a reaction while using the drug at home. All the children with positive DPT results developed mild cutaneous signs and presented with a reaction to amoxicillin-clavulanic acid. None had a systemic or severe cutaneous reaction. Conclusion: Initial direct single therapeutic dose challenge with a 5-day prolonged DPT is a useful and safe way to assess low-risk NIRs to BLAs in children.