Healthcare-associated infections in Bulgaria, 2003–2024: routine surveillance and repeated point-prevalence surveys

**Objectives:** To describe trends in Bulgaria's routinely reported healthcare-associated infections (HAIs) during 2003–2024 and compare them descriptively with findings from three European point-prevalence surveys (PPS). **Methods:** We conducted a retrospective descriptive study using aggregate national data. The annual routine indicator, defined as recorded HAI episodes per 100 hospital discharges, was summarised descriptively and modelled by ordinary least-squares regression with year coded 0–21. A sensitivity analysis excluded 2020–2021. Published PPS estimates and confidence intervals were reproduced; no inferential test compared PPS prevalence with the routine indicator because the measures have different designs and denominators. **Results:** The mean routinely reported HAI rate was 0.97 per 100 hospital discharges. It decreased from 1.48 in 2003 to 0.64 in 2024; the fitted average change was −0.0354 percentage points per year (R²=0.883) and was similar after excluding 2020–2021 (−0.0373; R²=0.898). In contrast, primary national PPS prevalence estimates were 3.7% (95% CI 2.8–5.0), 3.5% (1.7–6.8), and 3.7% (2.3–6.0). The PPS surveys included 42, 12, and 23 Bulgarian hospitals, with ECDC representativeness classified as optimal, poor, and medium, respectively. **Conclusions:** The two surveillance approaches produced discordant but non-interchangeable patterns. The discordance alone does not prove under-ascertainment, but it does warrant evaluation of case detection, reporting practices, sample composition, and changes in the surveillance system.