Patient representation in RNA signature studies for detecting and predicting the outcomes of febrile illnesses: a scoping review.

BACKGROUND: RNA signatures show promise for diagnosing and predicting outcomes in febrile illnesses. However, the demographic and geographic characteristics of patients included in RNA signature development studies was previously unclear, thus the generalizability of these signatures has been challenged.

METHODS: We conducted a scoping review of studies that identified blood RNA combinations for febrile illness diagnosis or prognosis. We searched EMBASE and Medline databases and extracted data on patient demographics, clinical settings, and geographic representation.

RESULTS: From 7,197 studies screened, 374 met inclusion criteria, yielding 286 unique RNA combinations (270 for infectious diseases, 16 for non-infectious causes). Most combinations targeted sepsis (n = 102), viral infections (n = 81), and tuberculosis (n = 51). Exclusively adult populations comprised 63% of study cohorts, while children only groups represented only 14%. 70% of studies recruited exclusively from inpatient settings. Patient groups from high-income countries accounted for 66% of infectious febrile illness patient groups. The greatest number of patient groups were from the United States, China, and the United Kingdom. Low- and lower-middle-income countries were significantly underrepresented despite bearing the highest infectious disease burden.

CONCLUSIONS: This scoping review shows that RNA signature studies for febrile illnesses have substantial demographic and geographic disparities compared to the real life disease burden. Research has focused predominantly on adult inpatients in high-income countries, potentially limiting the applicability of the resulting diagnostic tools for high-need, high-burden populations.

CLINICAL TRIAL NUMBER: Not applicable.