The aim of this study was to determine the accuracy of systemic inflammatory response syndrome (SIRS), quick Sepsis-related Organ Failure Assessment (qSOFA) score and GYM score to predict 30-day mortality in older non-severely dependent patients attended for an episode of infection in the emergency department (ED). We performed an analytical, observational, prospective cohort study including patients 75 years of age or older, without severe functional dependence, attended for an infectious process in 69 Spanish EDs for 2-day three-seasonal periods. Demographic, clinical and analytical data were collected. The primary outcome was 30-day mortality after the index event. We included 1071 patients, with a mean age of 83.6 [standard deviation (SD) 5.6] years; 544 (50.8%) were men. Seventy-two patients (6.5%) died within 30 days. SIRS criteria ≥ 2 had a sensitivity of 65% [95% confidence interval (CI) 53.1-75.9] and a specificity of 49% (95% CI 46.0-52.3), a qSOFA score ≥ 2 had a sensitivity of 28% (95% CI 18.2-39.8) and a specificity of 94% (95% CI 91.9-95.1), and a GYM score ≥ 1 had a sensitivity of 81% (95% CI 69.2-88.6) and a specificity of 45% (95% CI 41.6-47.9). A GYM score ≥ 1 and a qSOFA score ≥ 2 were the cut-offs with the highest sensitivity (p < 0.001) and specificity (p < 0.001), respectively. The area under the curve (AUC) was 0.73 (95% CI 0.66-0.79; p < 0.001) for the GYM score, 0.69 (95% CI 0.61-0.76; p < 0.001) for the qSOFA score and 0.65 (95% CI 0.59-0.72; p < 0.001) for SIRS. A GYM score ≥ 1 may be the most sensitive score and a qSOFA score ≥ 2 the most specific score to predict 30-day mortality in non-severely dependent older patients attended for acute infection in EDs.