Evaluation of an Internet-based decision-support system for applying the ATS/CDC guidelines for tuberculosis preventive therapy

Abstract

Preventive therapy for patients infected with tuberculosis (TB) remains an important component of TB control. To guide physicians in applying preventive therapy, the American Thoracic Society and Centers for Disease Control (ATS/CDC) developed guidelines based on PPD reactivity and on pretest probability of infection. The guidelines have become complex, and many clinicians find them challenging to apply. The authors developed a computerized decision-support system to assist clinicians in applying the ATS/CDC guidelines. This tool, published on the World Wide Web using hypertext markup language, delivers patient-specific recommendations based on physician-delivered patient-specific information. Four local TB experts derived eight TB infection scenarios and validated the web-based tool, which was tested for effectiveness using general internal medicine residents, randomly divided into two groups. Group A (n = 12) used the web-based tool and group B (n = 17) used pre-existing understanding of the guidelines and/or written resources to determine the need for preventive therapy in the case scenarios. Group A correctly used therapy in 92/96 possible cases (95.8%), group B in only 77/136 (56.6%) (p < 0.001). Group A required a mean of three mouse-clicks and 1.5 minutes per scenario to reach their choices, and they rated the web-based tool both intuitive and effective. These data demonstrate that a computer-based decision-support system for applying TB treatment guidelines can be delivered over the Internet and provide an efficient and effective resource for clinicians.