Improving the prognostication of lower respiratory tract infections in general practice

Lower respiratory tract infections (LRTI) are among the most common infections encountered in general practice with around three percent of adults consulting their general practitioner (GP) with an LRTI annually. Uncomplicated LRTIs generally run a favourable course. Nevertheless, adverse outcomes, such as hospitalisation or mortality, do occur. A prediction model to aid GPs in identifying LRTI patients at highest risk of adverse outcomes could improve clinical decision-making but is currently lacking in everyday practice. Evidence also consistently indicates that LRTIs can trigger acute cardiovascular events, such as myocardial infarction (MI), stroke, venous thromboembolism (VTE), and atrial fibrillation (AF). Although previous studies found that LRTIs increase the risk of cardiovascular events up to five-fold, none reported estimates of absolute risk.
This thesis focussed on improving the prognostication of patients presenting to general practice with LRTI. State-of-the-art prognostic and etiologic research methods were applied to electronic health records (EHR) data and it evaluated whether data derived from EHR unstructured clinical notes by natural language processing (NLP) improved prognostication.
In a systematic review on prognostic literature, increasing age, sex, current smoking, a history of diabetes, stroke, cancer, or heart failure, previous hospitalisation, influenza vaccination status, current use of systemic corticosteroids, recent antibiotic use, respiratory rate ≥25/minute, and a clinical diagnosis of pneumonia were identified as promising prognostic factors for hospitalization and mortality in LRTI patients in general practice. Currently available prediction models were considered not suitable for implementation in everyday clinical practice due to high risk of bias and incomplete assessment of model performance.
A model predicting individual risk of 30-day all-cause hospitalisation or mortality was developed using structured EHR data. The model included demographics, cardiometabolic diseases, other medical history, current medication use, and a clinical diagnosis of pneumonia. External validation yielded a c-statistic of 0.71 (95% CI 0.69–0.73), a calibration intercept of 0.28 (95% CI 0.20–0.36), and a calibration slope of 0.95 (95% CI 0.85–1.05). As such, the model holds promise to aid GPs in identifying LRTI patients at highest risk of 30-day hospitalisation or mortality.
Although large language models demonstrated good performance for automated extraction of signs and symptoms from Dutch EHR clinical notes, this information did not add substantial predictive value beyond structured EHR-based predictors alone.
Furthermore, excess cardiovascular risk following LRTI was quantified in this thesis. LRTIs were associated with an increased risk of major adverse cardiac and cerebrovascular events, VTE and new-onset AF, particularly during the first week following LRTI diagnosis. This culminates in approximately five to nine excess cardiovascular events attributable to LRTI per 1,000 LRTI patients. These cardiovascular risks warrant, at a minimum, increased awareness among clinicians that LRTIs are an undervalued risk factor for acute cardiovascular events. When developing interventions to mitigate cardiovascular risk in patients with LRTI, it is important to carefully balance benefits with potential risks and adverse effects to adhere to the “first do no harm” principle.