Enhancing Rural Healthcare in Emerging Countries through an eHealth Solution

Saatviga Sudhahar1, Dinusha Vatsalan1, Dulindra Wijethilake1, Yvonne Wickramasinghe1, Shiromi Arunathilake1, Keith Chapman2, Gihan Seneviratna1

1University of Colombo School of Computing, Sri Lanka 2Chilaw Base Hospital, Sri Lanka

Ministry of Health and Nutrition, Sri Lanka saatviga@gmail.com, dinush9@gmail.com, dulindraw@yahoo.com, yvonne.dilini@gmail.com,

sda@ucsc.cmb.ac.lk, keith_rpc@hotmail.com, gps@ucsc.cmb.ac.lk

Abstract— This paper investigates the applicability of an eSolution that improves healthcare delivery settings in the rural parts of the developing world and discusses about a proposed patient-centric eSolution that suits the Sri Lankan setting. Patients in rural areas incur heavy expenditure in travelling long distances spending lot of time to consult Specialists in cities due to the lack of Specialists in their areas. This issue can be addressed by an eSolution that makes appropriate use of Electronic Medical Records (EMR) and Telemedicine technologies which enables the patient to consult a Specialist through eConsultation. The main benefit of implementing such an eSolution is that, it ensures the availability of the Specialist across a distance at many eClinics within the shortest possible time frame. This paper also provides taxonomy of technology choices for connectivity and communication software for conferencing and discusses the overall solution that was proposed for use, considering its adaptability and appropriateness in a developing country perspective.

Keywords-eSolution; Electronic Medical Records; eClinic; eConsultation; eClinic Management Center

I. INTRODUCTION The Existing Health Sector in Sri Lanka consists of

around 10 tertiary level hospitals, over 27 secondary level hospitals and over 285 primary health care institutions. Healthcare delivery setting in rural parts of the island today exposes limited access to highly specialized consultancies. It is stated that 35% of the medical specialists in the curative sector are concentrated in the Colombo district [1]. Due to this patients in rural areas have to travel long distances to consult a Specialist in an urban area. This entails huge amount of cost, time and inconvenience especially in the elderly, post-operative and re-convalescing patients who have received specialized treatments.

Another interesting factor observed in hospitals in rural Sri Lanka is that, patient’s medical records are stored in the record rooms of the hospitals where they are ordered and managed manually. When a patient visits the clinic it is necessary to go through all the records to find the relevant record which imposes delay to the process. The following pictures show how patient’s medical records are kept in hospitals in Sri Lanka (Fig. 1) and how the patient records are piled up for searching a particular record when a clinic is about to start(Fig. 2).

Figure 1. Records kept in the record room at a hospital

Figure 2. Records kept in the Clinic for Search

This problem could be addressed by connecting the Patient and the Consultant through a web-based Electronic Medical Record (EMR) system along with a video-conferencing component through which a patient from his/her village peripheral hospital can consult a Specialist in the city/urban area with the assistance of a Doctor in an eClinic to receive treatment. The use of electronic medical records would also help manage patient’s medical records more efficiently.

This paper initially discusses similar systems that are used in other developing countries and highlights on the potential benefits that could be achieved through the proposed solution. It explains the pilot project which was carried out in Sri Lanka in order to address this problem and provides a detailed description of the technologies used and elaborates more on the system’s features. Finally the paper presents the importance of EMR security and confidentiality and concludes with a user study and the work that should be carried out in future.

2010 Second International Conference on eHealth, Telemedicine, and Social Medicine

978-0-7695-3950-8/10 $26.00 © 2010 IEEE

DOI 10.1109/eTELEMED.2010.11

22

2010 Second International Conference on eHealth, Telemedicine, and Social Medicine

978-0-7695-3950-8/10 $26.00 © 2010 IEEE

DOI 10.1109/eTELEMED.2010.11

23

II. RELATED WORK The Indian Space Research Organisation (ISRO) has

established a similar system in nearly 60 remote hospitals, which have been connected with 20 super-specialty city hospitals through satellite connectivity [14]. Though the system is user-friendly, people in the health sector lack knowledge in handling computers and other equipment and they also fear about losing their jobs.

In Tsilitwa, Africa an eClinic is connected to a doctor where Consultation happens through a video-link that is uni-directional. A voice call by the nurse is necessary to communicate with the doctor during consultation [15].

The National Referral Hospital in Honiara, Solomon Islands, has used a web-based system in Switzerland for telepathology consultations where doctors in Honiara submit cases and receive reports by email from a consultant [16]. The web-based system is used by consultants to discuss cases among an expert panel including remote pathologists. The problem here was that sometimes experts were not available to answer a request and Consultants were not provided with enough training which was necessary for proper collaboration in the system.

In Cambodia, a nurse travels from the Sihanouk Hospital Center of Hope to Markle Clinic once in a month where he examines photographs of patients and other vital statistics and sends a report to the hospital via email [5]. The hospital collaborates with other doctors on making a diagnosis. Delays are experienced in treatment and follow up of patients and the solution’s ongoing financial sustainability remains to be a challenge.

The proposed solution used a participatory approach in introducing the system where medical staff from both hospitals got involved in decision making from the initiation to the end of the project. Awareness programs and training programs were conducted for 6 weeks to doctors, consultants and nurses in both hospitals on Windows and Linux basics and also on usage of the EMR system. Real time communication between the consultant and patient eliminated delays in treatment and follow up and the solution was also built in a way that is cost-effective and sustainable from a developing country perspective.

III. EMR SYSTEMS IN DEVELOPING COUNTRIES Even as the developed countries are leading the electric

record revolution they are struggling to adapt to the new system, and at the current time only 9% percent of hospitals in the United States have adopted electronic medical record keeping. This figure is in stark contrast to the 60% of Indian Hospitals which are using electronic medical records keeping in their surgery rooms [11]. A few examples of EMR systems in developing countries are stated: • In Kenya, the MMRS and AMRS serve both clinical and

research needs. It supports comprehensive HIV care as well as mother-to-child transmission prevention while serving as a rich database for quality improvement [2].

• PIH-EMR in Peru to support the treatment of patients with MDR-TB. It collects baseline data on patient

assessment, bacteriology test results, drug regimens and patient outcomes [6].

• In the HIV-EMR system in Haiti, patient demographics, presenting complaint history, medications laboratory results, physical examination, social circumstances and treatment are recorded [7].

• In Malawi, Physicians, nurses and pharmacists perform all data entry using touch screens to the EMR system, including medication orders [8].

• In rural India, a Pocket PC-based system has been used to store health records for community nurses visiting remote villages [2].

• Iin KwaZuluNatal, South Africa, a Palm-based system allows secure access to HIV results in remote clinics [2].

IV. EMR VALUE CHAIN MODEL “Electronic medical record (EMR) is not only being

welcomed by healthcare providers as a way to improve care delivery but also serves as a catalyst and gold standard for development” (Porter, Kohane, & Goldman, 2005). The potential benefits that could be achieved by an EMR system are shown in Fig. 3 through a value chain model.

Figure 3. Value Chain Model for an EMR system in a developing country

The main benefit of implementing such a solution is the availability of the Specialist across a distance at many e-clinics within the shortest possible time frame. Patient’s travel expenditure, travelling time and the cost of transferring patients from peripheral hospitals to tertiary centers are drastically reduced. Specialists can treat patients in any part of the country giving wider access to patients all over the country. Decision making is easier since a holistic view of the patient’s medical information is available through the system. Medical reports serve the administrative purposes of the hospital. There is also a possibility of integrating the system with other systems and mobile interfaces. Knowledge transferring could happen between the Consultant and Doctor where the consultant can provide his ideas and suggestions to the Doctor while there is also a great potential for clinical research as data is stored electronically.

V. SCOPE OF THE PILOT PROJECT This pilot project was funded by the Information and

Communication Technology Agency of Sri Lanka (ICTA) and was initiated connecting the Marawila Base Hospital and

Primary Value

Processes

Support Value

Processes

Access to Patient information irrespective of time and place

Availability of

specialists across a

distance of eClinics

Easy Decision making with Holistic View of Patient information

Reduced cost and time

Clinical Research

Knowledge Transfer

Integration with other systems

Administrative Functionalities

2324

Dankotuwa Peripheral hospital which is situated around 33 miles away from Colombo in the North-western province of Sri Lanka. In the pilot project patients visited the Dankotuwa peripheral hospital from where they were able to consult the Specialist at the Base Hospital through the system. Initially patients from the Medical and Surgical clinics were considered. Later the Dermatology clinic was also included. Fig. 4 shows the scope diagram of the pilot project.

Figure 4. Scope diagram of the Pilot Project

The Consultant decides whether a patient is eligible to visit the eClinic at the peripheral hospital for the next consultation based on the criticality of the patient’s state. Patients from wards are also eligible to visit the eClinic for regular checkups once they are discharged from the hospital. Before attending the eClinic these patients should get themselves registered at the eClinic Management Centre (ECMC) at the Base Hospital, so that all their previous medical history and personal information would be updated to the system. From that onwards, they will be able to visit the eClinic for consultations.

VI. DEVELOPMENT OF THE EMR SYSTEM Since the system needs to communicate information between different locations over the internet using video conferencing, a web-based approach was adopted for development. Through the web based approach the system can be deployed in one machine and accessed by several others. Open Source technologies were utilized for development as it is cost-effective.

A. Existing EMR Systems in the Market There are so many EMR products in the market including

Open Source products similar to the proposed system. Due to

certain reasons, utilizing them was not appropriate. An evaluation of some of those systems is shown in Table 1. Most of the systems were stand-alone systems. Some involved huge complexity overhead in customization while some others had limited features that made them inappropriate for use. Performance and extreme cost issues were observed in a few systems.

B. Technical Feasibility The total tele-density (fixed and cellular connections) in

Sri Lanka is nearly 70, and also 100% fixed line connectivity is available to all the hospitals in rural and urban areas of the country [1]. ADSL connectivity was provided by the Sri Lanka Telecom (SLT) to both the hospitals with uplink and downlink speeds of 512kbps and 128 kbps.

Free communication software like Skype and Ekiga were used for video-conferencing by both the hospitals. Since Skype is being widely used by most of the Consultants in Sri Lanka for other communication needs, they were very comfortable in using it for conferencing. Skype is also more secure than today’s VoIP systems, since encryption is not part of most VoIP offerings. [3]. While Skype is mainly used for conferencing, Ekiga remains to be an alternative. Two computers are used together at the eClinic and the eConsultation center, where one is used to operate the web-based system and the other for conferencing.

Required hardware for the eClinic and eConsultation centers include, 4 workstations (or 2 workstation with dual screen capability) with onboard network cards & inbuilt speakers, UPSs, webcams, microphones, printers and a digital camera with 4GB memory card. The total cost incurred for the whole solution was approximately 400,000 LKR.

The system runs in Ubuntu 9.0 which is a Linux based operating system. Linux platforms are more secured compared to Windows based platforms because they are less vulnerable to viruses and trojan worm attacks. The technologies used for the development of the EMR system are shown in Table 2.

TABLE II TECHNOLIOGIES USED

Technology Backend EJB 3.0 & Hibernate Front end Struts FrameworkApplication Server JBoss 4.2.3 GA Build Tool Maven Database MySQL Community EditionEnterprise Java Beans (EJB) provides business level

services, transaction handling, security, and Remote Method Invocation (RMI) services to the application. Struts framework presents other frameworks like Tile and Validation frameworks, which makes development much easier and it also uses the MVC (Model-View-Controller) design pattern that provides reusability and maintainability of code.

2425

TABLE I EMR SYSTEM EVALUATION

VII. ARCHITECTURAL DESIGN

A. System Architecture The EMR system is based on Multi-Tier Architecture which involves the introduction of three software layers: The workstation where the user interacts with the program, the application server where data is calculated and processed and the database server where the data is stored. The system architecture of the web based system is shown in Fig. 5.

Figure 5. System Architecture Diagram of the Web-based System

With Multi Tier systems much of the data processing and business logic reside on the Application Server. The Application Server communicates directly with the Database Server, processes this information and then delivers the processed information to the Client. This architecture is scalable, can be controlled centrally and remote offsite backups can also be taken. For the pilot project the Sri Lanka Telecom (SLT) undertook responsibility for hosting the database and the application server at the SLT premises.

B. Application Design The EMR application consists of five major components.

They are Patient Management, Scheduling, User Management, Reporting and My Profile.

1) Patient Management

Initially patients are registered through the system and they are provided with an eClinic Id through which the system tracks all patient records. Patient Management manages the medical records of all patients registered to the system. The details in each patient’s medical record are divided in to five areas. They are details regarding the patient’s Demographics, Clinical Chart, Investigations, Patient Images regarding the complaint and the recorded Audio/Video conversations between the Consultant and the Patient.

Demographic details consist of the patient’s personal information, the clinic from which he/she was referred and the hospital where the patient has received treatment. The Clinical Chart consists of the information regarding the clinics to which the patient has attended before. According to the suggestions from Consultants/Doctors, clinical information was further divided in to sub-categories. They are the patient’s, Presenting Symptoms and Complaints, Past Medical History and Social History, Family History, Allergies, Examination details, Diagnosis/Provisional Diagnosis, Treatment, Follow up Plan and Summary.

The consultant and doctor will be able to view the previous patient information under each of these categories that is sorted in a way where the lastly entered data is shown in the top. Information entered under these categories by the Consultant and the Doctor would be updated to the record with the Date, Time and the Doctor’s name who has entered it. During this process the Consultant will be having a conversation with the Doctor and patient simultaneously through video-conferencing. The treatment and follow up plan will be updated by the Consultant and prescription details will be given to the patient after the Consultation. An image of the ongoing video-conferencing and treatment through the eClinic at Dankotuwa is shown in Fig. 6.

Investigations consist of the results of investigations undergone by patients. Full Blood Count, Fasting Blood Glucose, Ultra filtration Rate, Urine Culture, Erythrocyte Sedimentation Rate, Fluorescent Antibody Test, Thyroid etc are some of the reports that have been incorporated up to

EMR System Architecture Licence CommentsFileMed EMR Standalone Proprietary Simple and easy to use, strong security[9]Patient OS Standalone GNU (GPL) customization is complicated Hospital OS Web-based GNU (GPL) Focuses on hospital information, Suitable for small-

sized hospitals [10]Open EMR standalone GNU (GPL) Limited Features Open MRS Web-based OpenMRS Public Licence

Limited Features [18]

Medical Web-based GNU (GPL)

Linux based , Industry Standards, Too slow

GNU Med Web-based Free in a special way, modified code should be patented

Complicated, Limited Features, Too slow [12]

OSCAR EMR Web-based GNU (GPL) Hardware costs, network support costs high [13]

2526

now. The most significant details in a report are highlighted according to the date on which the investigation was done.

Figure 6. Patient and Doctor during Consultation at the eClinic

Patient images section provides with all the images of the

patient regarding the illness or complaint. The images can be taken by a digital camera by the doctor and can be uploaded to the system. Radiology and ECG images are captured by camera and uploaded to the system which is still very precise. Images are updated with the image annotation, uploaded date and time and the patient referred date. The screenshot of the patient images in the system is shown in Fig. 7.

Figure 7. Screenshot of Patient Medical Images in the system

Audio/Video conversations between the Doctor,

Consultant and the patient are recorded and uploaded to the system as well.

2) Scheduling

Scheduling deals with creation, rescheduling and cancellations of eClinic Appointments and Theatre Schedules through the system. When the consultant logs in to the system he will be able to view the eClinic Appointments /Theatre Schedules for the day with the patient’s name and eClinic Id in the home page.

3) User Management

The most critical feature in this component is Role-based User Access Privileges. Access privileges provide a

level of security, preventing users from tampering with aspects of the in which system they don't belong and allows a system to provide customized interfaces to users based on their roles. The system deals with five categories of users namely, Consultants, Doctors, Medical Superintendents, Nurses and System Administrators.

These user categories have different access privileges to the system according to their roles. Consultants and Doctors have the privilege of viewing adding and editing patient details while Nurses and Medical Superintendents have only the read privilege. Currently access privileges are provided in the system for each area of the patient’s record like, view/edit/add privilege for demographics, clinical chart, investigation, patient images and the audio/video conversations.

4) Reporting

Reporting was implemented for administrative purposes where monthly Appointment Reports and Theatre Schedule Reports can be obtained. They are dynamically generated at real time.

5) My Profile

Personalization of user profiles is provided through this section.

VIII. SECURITY AND CONFIDENTIALITY The adequate protection of patient health record requires

limitations at all levels, i.e. collection, use, access, and disclosure. Patients can face serious risks if their communities discover their illness or other sensitive medical information. Well-run projects should use a combination of technical and human protocols.

Details relevant to the patient’s medical record cannot be deleted by anyone once it is updated in the system but can only be viewed or appended. Users are required to have complex passwords and change passwords regularly. Only relevant access privileges should be applied to each user category based on their roles. All log-ins and viewed data should be recorded in a system log and reviewed to minimise the occurrence of unauthorized access. A centralised database system should be used to allow the computer and data to be physically secured and backed up on a regular basis. Encryption of data transfers should be performed using the Secure Socket Layer (SSL) protocol. Database review and report generation for purposes other than direct care should be restricted [4]. It is also imperative to be in compliance with the Site Disclaimer, Privacy Policy and the Terms and Conditions on the information in the system.

IX. USER STUDY According to the surveys conducted from patients

visiting the eClinic, there was a significant reduction in total travel cost (90.94%), in travel distance (80%) and travel time (75%) after the solution was adapted to the hospital system. Consultants and Doctors were more enthusiastic

2627

about it and they were willing to provide their contribution for the future development of the system.

X. FUTURE WORK Currently the time taken for an eConsultation per patient

is approximately 7-10 minutes. This figure is high compared to the average time taken in normal consultations. According to the suggestions by Consultants it can be further reduced if any speech recognition or handwritten recognition software is used. The ability of the recognition software to identify the acronyms used by Consultants and Doctors correctly is still a question. It would be a critical issue if a word has been mistakenly interpreted by the software. The system could be further integrated with equipments like ECG machines and X-ray scanners, so that these images could be obtained much more precisely. Incorporating standards like HL7 and DICOM is subject to future work. Accessing medical records through mobile interfaces using hand-held devices will also be future work. Since most of the consultants carry PDAs or high-end mobile devices implementing that would be more convenient for them. Uploading images and audio/video conversations to the system requires high-speed internet connectivity. Research has to be done on the usage of appropriate compression techniques with optimized protocols in order to increase the speed of data transfers.

XI. CONCLUSION This paper presented a detail description of the eSolution

as a proof of technology which was developed to support the problems prevailing in the current healthcare settings in developing counties. It discussed the significance of implementing an Electronic Medical Record system and then talked about the system architecture, application design, technologies used and how the system model was optimized to suit the conditions in a developing country considering the critical issues, challenges and security constraints as well. It is concluded that this is a promising solution that facilitates health consultations in rural communities of developing Countries with less Cost, Travel Time and Travel Distance, making a healthy population that will in turn catalyze the global development process.

ACKNOWLEDGMENT I would like to thank Dr. N. Nugaduwa, Consultant

Surgeon, Consultants and Doctors of the Marawila Base Hospital and the Dankotuwa District Hospital, Senior academic staff of University of Colombo School of Computing (UCSC) and staff of the Advanced Digital Media Technology Centre, UCSC. My gratitude is extended to the Information and Communication Technology Agency of Sri Lanka (ICTA) who funded the pilot project and Sri Lanka Telecom (SLT) who provided the connectivity between the hospitals.

REFERENCES [1] Chapman, K.R.P. and Arunathilake S.M. K.D., “Introducing a

Patient Centric e-Health Solution for a Developing Country”, presented at the 2009 Int Conf. IADIS eSociety, Spain.

[2] Fraser HS, Biondich P, Moodley D, Choi S, Mamlin BW, Szolovits P. (2005). Implementing electronic medical record systems in developing countries. Informatics in Primary Care. [Online]. 13. 83-95 Available: http://groups.csail.mit.edu/medg/ftp/psz/EMR-design-paper.pdf [accessed: Jan 2, 2009]

[3] Simson L. Garfinkel (2005). VoIP and Skype Security. [Online].Available: http://www1.cs.columbia.edu/~salman/skype/OSI_Skype6.pdf [accessed: Jan 11, 2009]

[4] Gary Kurtz and FHIMSS. (2002, Dec.). EMR Confidentiality and Security. Journal of Healthcare Information Management. [Online]. 17(3). 41-48. Available: http://www.providersedge.com/ehdocs/ehr_articles/EMR_Confidentiality_and_Information_Security.pdf [accessed: Dec 30, 2008]

[5] Lygeia Ricciardi,( 2004). A Model for Remote Health Care in the Developing World:The Markle Foundation Telemedicine Clinic in Cambodia,[Online]. Available: http://www.connectingforhealth.org/resources/telemedicine_clinic.pdf [accessed: Jan 28, 2009]

[6] Partners in Health. PIH-EMR Overview.[Online]. Available: http://model.pih.org/electronic_medical_records/pih_emr_overview [accessed: Mar 3, 2009]

[7] BMJ. (2004, November) Information in practice. An information system and medical record to support HIV treatment in rural Haiti. [Online]. Available: http://www.bmj.com/cgi/content/full/bmj;329/7475/1142 [accessed: Feb 20, 2009]

[8] Douglas G. The Lilongwe Central Hospital Patient Management Information System. Proceedings of the AMIA Annual Fall Symposium 2003;833

[9] Filemed. Electronic Medical Records (EMR) and Appointment Scheduling Software. [Online]. Available: http://www.filemed.com/ [accessed: Jan 18, 2009]

[10] Sourceforge. Hospital OS. [Online]. Available: http://sourceforge.net/projects/hospital-os/files/ [accessed: Feb 10, 2009]

[11] OpenEMR Home Page. [Online]. Available: http://www.oemr.org/ [accessed: Feb 27, 2009]

[12] Jack Wallen.(2009). Let your medical practice go open source with Gnumed. [Online]. Available: http://www.ghacks.net/2009/10/20/let-your-medical-practice-go-open-source-with-gnumed/ [accessed: April 1, 2009]

[13] Oscar. Brief Overview. [Online]. Available: http://www.oscarcanada.org/about-oscar/brief-overview [accessed: Mar 2, 2009]

[14] Bagchi S (2006) Telemedicine in Rural India. Public Library of Science. [Online]. 3(3). Available: http://www.plosmedicine.org/article/fetchObjectAttachment.action;jsessionid=5C8684F6106A7B8F4D32A377BA1D406B?uri=info%3Adoi%2F10.1371%2Fjournal.pmed.0030082&representation=PDF [accessed: Feb 20, 2009]

[15] Tele-Health application demonstrated over wireless network in Tsilitwa and Sulenkama, Eastern Cape.[Online]. Available: http://www.cda.co.za/Tsilitwa/Events/TeleHealth%20Appl/TeleHealth.html [accessed: Jan 2, 2009]

[16] K Brauchli, R Jagilly, H Oberli, K D Kunze, G Phillips, N Hurwitz and M Oberholzer (2004). Telepathology on the Solomon Islands. Journal of Telemedicine and Telecare. [Online]. 10 (Suppl. 1). 14–17. Available: http://ipath.ch/site/files/Brauchli_JTT04_p14.pdf [accessed: Mar 2, 2009]

2728