BACKGROUND The assessment of airway responsiveness and inflammation is key to the investigation of occupational asthma (OA). OBJECTIVE We sought to assess and compare the diagnostic accuracies of the blood and sputum eosinophil counts and the methacholine challenge for the diagnosis of OA. METHODS We conducted a retrospective study assessing 618 patients who underwent specific inhalation challenges (SICs) for symptoms suggestive of OA between 2000 and 2015. A sputum induction and a methacholine challenge were performed before and after SICs. Blood samples were collected in all subjects before the SICs and in 100 subjects before and after SICs. The diagnostic accuracies of blood and sputum eosinophil counts and methacholine challenge were calculated for diagnosing OA. RESULTS The change in blood eosinophil count failed to differentiate workers with positive and negative SICs. The change in sputum eosinophil counts induced by the exposure to the offending agent had the highest diagnostic accuracy (receiver operating characteristic area under the curve: 86% [95% confidence interval: 0.8-0.9, P < .001]) for diagnosing OA compared with changes in concentration of methacholine inducing a 20% fall in forced expiratory volume in 1 second (PC20) and blood eosinophils. Combining a 2-fold or greater decrease in PC20 or a 3% or greater increase in sputum eosinophil count achieved a sensitivity of 84% and a specificity of 74% with a negative predictive value of 91% for the diagnosis of OA. CONCLUSIONS Blood eosinophil counts do not appear to be an effective aid for diagnosing OA. The performance of both sputum cell count analysis and a methacholine challenge before and after exposure to the offending agent may represent an effective alternative in diagnosing OA when SICs are unavailable.