Project Details

Using a mobile application to improve pediatric presumptive TB identification in western Kenya

Daria Szkwarko, DO, MPH

Summary

BACKGROUND: Early recognition of TB symptoms in children is critical in order to link children to appropriate TB treatment and decrease complications. Healthcare workers in pediatric outpatient clinics in limited resources settings like Webuye County Hospital in western Kenya are often overburdened with competing clinical priorities, leading to incomplete symptom screening for presumptive TB.
 GAP: The implementation of screening tools such as a presumptive pediatric TB mobile/tablet application show promise; however, although this tool has been implemented for care in both Bangladesh and Kenya, it has never been formally evaluated for feasibility, appropriateness, and effectiveness. HYPOTHESIS: We hypothesized that the implementation of this user-centered designed presumptive pediatric TB mobile/tablet application will be feasible and appropriate in our setting. We hypothesized that the proportion of presumptive TB patients and pediatric TB cases identified in outpatient clinics will significantly increase with the use of the presumptive pediatric TB mobile/tablet application. METHODS: Using a mixed-methods implementation science framework, we used a participatory, iterative approach to pilot and adapt the presumptive pediatric TB mobile/tablet application based on feedback from healthcare workers in pediatric outpatient clinics in a county hospital in western Kenya. We reviewed data for children < 15 years who attended pediatric outpatient clinic, nutrition clinic, casualty, child welfare clinics and other pediatric outpatient areas before and after the implementation of the mobile application, and we compared the number of children identified in the paper presumptive TB registers. RESULTS: Between August 2019-January 2020, 1787 children age ≤15 were screened for presumptive TB using the PPTBMAPP, 376 (21%) of which met the criteria for presumptive TB. Baseline characteristics available in the presumptive TB registers included sex, HIV status, and age. The baseline characteristics between the pre- and post-mobile application groups were similar except for age; the pre-mobile application group was significantly older than the post-mobile application group (median age in months, IQR: 120, 84-144 and 48, 22-120 respectively with P<0.001, table 1). In the presumptive TB registers, there was a statistically significant increase in the proportion of children to all patients between the pre-and post-mobile application periods (97/908 (10.7%) vs. 160/989 (16.2%) respectively, p=0.0005). In the active TB register, there was an increase in the proportion of children to all patients between the pre-and post-mobile application periods (17/117 (14.5%) vs. 15/83 (18.1%) respectively, P=0.5), but this was not statically significant. According to the key informant interviews, 6 out of 7 healthcare workers interviewed commented that the application sped up the presumptive TB screening process and that the program would likely lead to increased pediatric TB diagnosis. One healthcare worker stated, “I think the CHV has done a very wonderful job. We can capture more cases. And that is why our number of pediatric numbers has increased.” IMPACT: This study provides preliminary data regarding the effectiveness of a CHV led presumptive pediatric TB mobile application, thereby laying the groundwork for future research to test this mobile application and identify an evidence-based strategy to ultimately decrease the burden of pediatric TB globally.

Publications: