Operational Assessment of Laboratory Information System for MDR-TB in Lima, Peru

Overview

The objective of this study is to compare the effects of a web-based laboratory information system (e-Chasqui) between a network of health establishments with access to e-Chasqui (intervention group) and a network of health establishments without access to e-Chasqui (control group). The specific aims are: 1. To compare the "laboratory turn-around-time" (from the date a culture or drug susceptibility test (DST) result is obtained to the date the result is obtained at the health center) of samples pertaining to health establishments in the intervention versus the control group. 2. To compare the "clinical turn-around-time" (from the date the DST result is obtained to the date the patient is evaluated by a physician in possession of that result) among multi-drug-resistant tuberculosis (MDR-TB) patients pertaining to health establishments in the intervention versus control group. 3. To compare the laboratory reporting errors (defined as incorrect smear, culture, or DST results) between health establishments in the intervention versus control group. 4. To qualitatively assess the acceptability and usability of e-Chasqui among users in health establishments with access to the system. The investigators aim to test the following hypotheses: 1. The laboratory turn-around-time for health establishments with e-Chasqui access will be smaller than that for establishments without e-Chasqui access. 2. The clinical turn-around-time for patients pertaining to health establishments with e-Chasqui access will be smaller than that for patients in establishments without e-Chasqui access. 3. Health establishments with e-Chasqui access will have fewer errors compared to those without e-Chasqui access. 4. Factors associated with acceptability and usability of e-Chasqui by systems users can be identified.

Delays in starting patients with multi-drug resistant tuberculosis (MDR-TB) on appropriate medication treatment lead to worsened patient outcomes and increased risk of transmission. In Peru, the National Tuberculosis Program (NTP) has provided treatment for patients with MDR-TB since 1996, with cure rates ranging from 48% to 83%. Nonetheless, significant delays exist in the diagnosis of MDR cases and timely treatment initiation, in particular since MDR-TB treatment has been scaled to a national level. Our preliminary work has found that the average turn-around-time (TAT) from the request of a DST to the prescription of appropriate treatment was 148 days (4.9 months). An information system could be used to virtually eliminate part of this delay: the time to communicate results between the different institutions. In 2001 we developed and deployed a web based medical record system, the PIH-EMR, to assist in the clinical management of these MDR-TB patients. This system includes a substantial set of data entry and analysis tools for laboratory results including sputum smears, cultures and DSTs. Over the last six months the PIH-EMR has been modified to support the decentralization of DSTs in the regional laboratories. This laboratory component of the PIH-EMR, termed "e-Chasqui," permits web-based entry of culture and DST results at all regional and central laboratories performing these tests. In addition, the system includes applications to assess quality control, generate aggregate reports, notify health centers of new results or contaminated samples, and track enrolled patients and the status of pending laboratory tests. The NTP and the National Reference Laboratory (NRL) have agreed to integrate e-Chasqui into the current management of patients at risk of MDR-TB in two regions, and if effective, to provide a national network of all laboratories performing DST. Because of the intensive initial demands on resources and training, the implementation of the information system will occur in stages (parts A and B of this study). Broadly put, this study will provide data on whether the use of information systems can be shown to improve quality of patient care. More specifically, it will assess the impact of e-Chasqui on reducing reporting delays and laboratory data errors, and improving patient outcomes. This is a prospective observational controlled study to compare the effects of a web-based laboratory information system (e-Chasqui) between a network of health establishments with e-Chasqui access (intervention group) compared with a network of health establishments without e-Chasqui access (control group). The data is being collected by the overarching study "Operational Assessment of Diagnostic Methods for MDR-TB in Lima, Peru." The study is planned in two parts: Part A will be a simultaneous control study occurring in the "transition" period comparing endpoints in those health establishments with e-Chasqui access to those in matched health establishments without e-Chasqui access. Part B will be an historical control study comparing endpoints in the establishments which gain initial e-Chasqui access, before (historical control) and after (intervention) the implementation. Data sources include: patient charts, interview with health providers to confirm or clarify accuracy of data in patient charts, microbiology registries at local, regional and national laboratories, and the information system database. The total number of subjects to be enrolled study-wide is 1849.