الفهرس 
TitleOnly 14 pages are availabe for public view
 |  | from | 87 |  |  |
| | | from | 87 | | |
Abstract
Background: Infant and child mortality is unacceptably high in developing countries. In 2015 about 5.2 million children died before their fifth birthday in low and lower-middle income countries (6, 22). IMCI was developed to address the most common causes of morbidity and mortality among children in developing countries aiming at improving the diagnosis and treatment of common causes of childhood morbidity and mortality at the primary level of health care (28). However, the extensive paperwork and the time it takes to follow the IMCI chart booklet extends the examination time which means longer waiting time, overcrowding, and dissatisfactions among mothers/caretakers. There are also other barriers against adopting IMCI such as the training cost, frequent health worker turnover and poor quality of IMCI implementation by those trained (37). Most of these barriers could be addressed using clinical decision support systems.
Methods: Rules are extracted from the IMCI chart booklet and interviews with experts in IMCI. Later on rules were integrated in a hybrid mobile application. The validation of the decisions made by these rules were investigated by experts from MOHP. On the last stage, a RCT was done to evaluate the effect of the new knowledge source over the service delivery time. Four clinical scenarios were obtained from MOHP. Twelve physicians were randomized into two equal groups using a list generated from randomizer package in R. One group was instructed to use their conventional methods (IMCI chart booklet) (pIMCI) and the other was instructed to use the application (eIMCI) to solve the exact four cases. The duration was recorded and the resulting medical records were collected. At the end of the trial, the pIMCI group were given the chance to try the application. All the twelve participants were asked to fill SUS questionnaire and to do a short interview to collect their feedback.
53/60
Results: Participants of eIMCI group were faster than pIMCI by an average of 10 minutes per case (p = .002). The application got an average of 87.1 on the SUS which means “Excellent user experience”. The eIMCI showed 100% complete records and because of the decision making automation, the subsequent decisions were 100% accurate and complete. The cases in pIMCI (n=24) showed 16 cases (66.7%) with incomplete treatment, 12 cases (50%) with missing doses a 5 cases (20.8%) with miscalculated doses, 14 cases (58.3%) with inaccurate immunization decisions, 4 cases (16.7%) were misclassified, 4 cases (16.7%) were unnecessarily referred to the hospital while two cases out of six (33.3%) who were supposed to be referred to the hospital didn’t. Physicians showed poor quality medical records with incomplete recording of signs and symptoms in 19 cases (79%).
Conclusion: The eIMCI is faster than pIMCI and more accurate. The medical records are better regarding its completion and structure quality. The application showed excellent acceptance from the target users.