Simulation and visualization of large scale distributed health system infrastructure of developing countries

Abstract

Developing countries are faced with a number of health-care challenges: long waiting hours of patients in long queues is just one of such challenges. The key cause of this has been identi ed to be a lack, or uneven distribution human resources among health facilities. This sets the stage for poor and ine cient delivery of quality primary health care, especially to the rural dweller as they usually have a fewer medical professionals in their area. The impact of this a ects not only the state of health of the population, but also the economy, and population growth of the a ected community. To try and address this, the introduction of Information Technology (IT) into health-care has been suggested by many health governing bodies like theWorld Health Organization (WHO) and other authorities in health care. The ability of IT to go beyond physical boarders and extend professional care has been the key characteristic that supports its integration into health-care. This has eventually lead to the development of Health Information Systems (HIS) that support remote consultation. Despite all these innovations, there is still evidence of poor and ine cient delivery of services at health facilities in many developing countries. We propose a completely di erent approach of addressing the problem of poor and ine cient delivery of health-care services. The key challenge we address is that of lengthy queues and long waiting hours of patients in health facilities. To cut down on the use of nancial resources (whose lack or shortage is a major challenge in developing economies), we propose an approach that focuses on the routing of patients within and between health facilities. The hypothesis for this study is based on a suggestion that alterations to the routing of patients would have an e ect on the identi ed challenges we seek to address in this study. To support this claim, a simulator of the health system was built using the OMNET++ simulation package. Analysis of test-runs for di erent scenarios were then tested and the simulation results were compared against controls to validate the functioning of the simulator. Upon validation of the simulator, it was then used to test the hypothesis. With data from the di erent health-care facilities used as input parameters to the simulator, various simulation runs were executed to mimic di erent routing scenarios. Results from the di erent simulation runs were then analyzed. The results from the simulator and analysis of these results revealed that: In a case where patients where not given the liberty to consult with a doctor of their choice but rather to consult with the next available doctor/specialist, the average time spent by patients dropped by 26%. The analysis also revealed that moving a receptionist from the rst stage upon patient entry into the health facility reduced the average patient life time by 85%. This was found to be a consequence of a drop in queue length (a 28% drop in queue length). On the other hand, the analysis also revealed that the total removal of a general receptionist increased patient life-time in a facility by 30.19%. This study also revealed that if specialists were deployed to certain health facilities rather than having referred patients come to them in the urban health facilities, patient population in the urban health centers will drop by 32%. This also saw a drop in patient waiting time in the rural health centers as more doctors were available (a reduced patient-to-doctor ratio in rural health facilities). The results from the analysis support our hypothesis and revealed that indeed, alterations to the way patients are routed does have an e ect on the queue lengths and total waiting time of patients in the health system.