Incidence, predictors, and prevention practices of ventilator-associated pneumonia among critically ill patients at the critical care unit, Kenyatta National Hospital, Kenya: A retrospective mixed-methods cohort study

Abstract

Background: Ventilator-associated pneumonia (VAP) is among the most serious healthcare-associated infections encountered in intensive care units worldwide. In low- and middle-income countries, Kenya included. This study determined the incidence, predictors, and preventive practices of ventilatorassociated pneumonia (VAP) among mechanically ventilated patients at the Critical Care Unit (CCU),
Kenyatta National Hospital (KNH), KENYA.

Methods: A retrospective mixed-methods cohort study design. The quantitative strand involved retrospective cohort analysis of 197 mechanically ventilated adult patient records between June 2024 and December 2024. Patients included had no respiratory infection at intubation. VAP was determined using an adopted modified percentage-based Clinical Pulmonary Infection Score (CPIS) threshold of ≥40% to confirm diagnostic variability. Predictor variables included demographic, clinical, physiological, and CCU care characteristics, were tested using univariate and multivariate analysis. The qualitative strand comprised three focused group discussions (FGDs) made of 13 critical care nurses, and the analysed using Braun and Clarke's reflexive thematic analysis.

Results: Cumulative incidence of VAP was 37.%(73 of 197 patients). Multivariable analysis revealed two independent predictors: white blood cell (WBC) count ([aOR] = 1.85; 95% CI 1.57–2.18; p < 0.001) and haematological malignancy as the primary admission diagnosis (aOR = 15.32; 95% CI 2.99–78.43; p < 0.001). These predictor model showed excellent ability to discriminate VAP from non-VAP patients. The female gender(p=0.013), lower age(0.022) and lower weight(p=0.048) were associated with VAP on univariate analysis. Qualitative analysis yielded five themes: perceived decline in VAP burden; recognition of patient-level risk; routine application of prevention bundle elements; systemic failures; and gaps in formal competency assessment.

Conclusions: VAP incidence at KNH has declined from 54.4% in 2018 to 37.1% in 2024 but remains highe. Residual risk appears to be driven by patient biological vulnerability rather than deficiencies in routine preventive care. Risk-stratified prevention protocols, infrastructure investment, and staff structured competency assessment is recommended.