Why is In-hospital Diagnosis of Pulmonary Tuberculosis Delayed in Southern Taiwan?

Background/Purpose: In-hospital diagnosis delay (IHDD) of pulmonary tuberculosis (TB) has a significant impact on nosocomial TB transmission. We investigated the risk factors associated with prolonged IHDD in Taiwan, a high-resource, mid-incidence area. Methods: Between January 2005 and August 2006, we retrospectively enrolled 193 consecutive hospitalized patients. All of them had culture-proven pulmonary TB and did not receive antitubercular treatment at admission. IHDD was defined as the interval between admission and initiation of antitubercular treatment. Patients were grouped according to the median value of IHDD. Results: The median IHDD was 7 days. Patients with IHDD > 7 days were considered the prolonged-delay group, and those with IHDD <= 7 days, the short-delay group. Independent risk factors [with adjusted odd ratios (95% confidence intervals)] for prolonged IHDD were: negative sputum smear [47.53 (13.20-171.18), p<0.001]; non-cavitary lesions on chest radiographs [14.90 (3.46-64.14), p<0.001]; admission to hospital departments other than chest medicine/infectious diseases [6.60 (1.95-22.41), p=0.0021]; exposure to fluoroquinolones before antitubercular treatment [5.29 (1.13-24.75), p=0.034]; underlying malignancy [4.59 (1.13-18.67), p=0.033); and age > 65 years [3.19 (1.01-10.05), p=0.048]. Death attributed to tuberculosis was associated with positive sputum smear (hazard ratio = 21.85; 95% Cl=2.74-174.44; p=0.004) but not prolonged IHDD (p=0.325). Conclusion: To minimize IHDD, clinicians should carefully manage hospitalized patients with risk factors for prolonged delay, such as those with negative sputum smears, non-cavitary lesions on chest radiographs, admission to departments other than chest medicine/infectious diseases, exposure to fluoroquinolones before antitubercular treatment, underlying malignancy, and age > 65 years.