Diarrheal diseases lead to an estimated 2.4 billion episodes of illness and 1.3 million deaths each year, with the majority of those deaths occurring in adults, adolescents, and children over five years. As the severity of diarrheal diseases can vary widely, accurately assessing dehydration status remains the most crucial step in preventing morbidity and mortality. While patients with severe dehydration require hospital admission and immediate resuscitation with intravenous fluids to prevent hemodynamic compromise, organ ischemia, and death, those with mild to moderate dehydration can be treated in outpatient settings with relatively inexpensive oral rehydration solution. Yet, while several tools have been validated for use in children under five years of age, no clinical diagnostic tool has ever been validated for the assessment of dehydration severity in adults, adolescents or children over five years of age with acute diarrhea. Differences in both adult physiology and diarrhea etiology may compromise the accuracy of clinical diagnostic models developed for use in young children. The proposed research will derive the very first age-specific clinical diagnostic models created for the assessment of dehydration status in patients over five years of age with acute diarrhea, incorporate those models into a new mobile health (mHealth) tool, and validate the performance of this tool in a new population of patients with acute diarrhea.
To accomplish this task, we will enroll a prospective cohort of adults and children over five years of age with acute diarrhea presenting to the rehydration unit of the International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) in Dhaka, Bangladesh and collect data on presenting clinical signs and symptoms shown to correlate with dehydration severity in prior studies. We will then employ machine learning techniques to derive age-specific clinical diagnostic models for assessing dehydration in patients over five years of age with acute diarrhea. We will conduct formative research among clinicians working at icddr,b to develop an innovative mobile phone based platform which will incorporate these new age-specific diagnostic models for rapid use by frontline health workers. Finally, we will validate both the accuracy and reliability of the newly developed mHealth tool in a new population of adults and children over five years of age with acute diarrhea. Once developed and properly validated, this novel mHealth tool has the potential to help physicians, nurses, and other healthcare providers more accurately diagnose dehydration severity and better determine the optimal management strategy for patients with acute diarrhea. Improved diagnostic approaches may in turn be shown to reduce both the morbidity and mortality that occurs as a result of missed diagnoses of dehydration, as well as the adverse events and inappropriate utilization of limited healthcare resources that can result from inaccurate diagnoses of dehydration.