OpenEMR Success Stories

Overview

Plan to place clinic success stories here. Can be entries such as:

• A review type entry (a snapshot in time) on this page
• A link to an article/blog/sourceforge review/case study entry
• A link to a blog that describes the deployment over time

We are a start-up in the Bay Area of California called MediGrail LLC. We think of our company as having a
profit and a non-profit side. The OpenEMR implementation described below is a part of the latter. 
*  Application Custom reports have been developed for the user and SnoMED and ICD 9 databases are
installed. The current project is converting to ICD 10 codes and is experiencing some issues with it.
Populated the pharmacy and are printing patient payment receipts. Populated the pharmacy database with
inventory. It is tracking inventory. Auto-incrementing patient ID is also configured. The application team's
biggest challenge was to define ACLs for each department and their heads.
*  Goal The whole team including management has been very clear on the overall goal that everybody is
working towards. Last year the facility saw roughly 120,000 patient encounters and the goal of the automation
is to double that!
*  Team-building We set user expectations, provided role definitions, and helped in team formation.
*  Training The user-base is not trained in computing technologies other than extensive use of cell
feature phone SMS. The support team at Siaya also needs formal systems and applications team. Before going
live, the application team engaged in "Train the Trainers" (TOTs) activity. The TOTs then engaged in training
the staff while the support staff made themselves available to answer questions.
*  Installation time-frame December 2011 to January 2012
*  Location Western, rural Kenya and the government facility is called Siaya District Hospital
*  Contact Dr. Omoto Jackton, telephone +254 -721-761484, Yudhvir Singh Sidhu, sidhu@medigrail.com
*  Address  Medical Superintendent, Siaya District Hospital, P. O. BOX 144 Siaya, Kenya, Postal Code 40600
*  When Siaya District Hospital goes live 13 April 2012
*  What Automated a 220-bed hospital - network, workstations, servers, and UPS
*  Who Created a team split in 4 groups: the systems administrators, Facilities manager, IT manager, and
applications administrators. This team consists of Siaya people and the MediGrail staff members. 
*  Equipment  
*  An Intel Atom Supermicro server with 4 GB RAM and a 32 GB SSD drive which is the firewall and a Linux boot
server - running pfSense (pfsense.org)
*  An Intel Atom Supermicro server with 4 GB RAM and a 2 TB drive which is the file server, IM server and
future inter-department phone switch running on Ubuntu Linux and mounting user directories via sshfs.
*  Intel Xeon Sun server with 6 GB RAM and two drives which is the application server during the day and
backup server at night. The application is ofcourse OpenEMR (open-emr.org) and bacula.org for backups. 
*  The backbone is provided by HP Procurve 9078a 24-port Gigabit switches connected via Fiber Optic cables
*  Currently 40 Panasonic toughbook laptops
*  Inverting Opti-ups UPS and APC UPS
*  Network The systems administration team laid over 9,000 feet of network cabling in conduit and
terminated all the endpoints. Installed 3 switches and connected them via Fiber Optic cable runs. The network
will expand to 5 switches and each will be protected by voltage regulators and UPS.
*  Workstations We had to change our original thin-client stand-alone workstation design to laptops
because you cannot reliably ship equipment to Kenya. The equipment went with us on the airplane. The
workstations have a minimum of 512 MB RAM, no hard drive and no battery. This is intentional given the
sensitivity of the data. Laptops boot off the network into a very small 55 MB footprint Linux distribution
called Slitaz. It contains Abiword, Gunmeric spreadsheet, Firefox browser, a PDF viewer, and a file server
mount utility called sshfs.
*  Power Creating a stable and clean power source is a major challenge at the facility and one which we
are currently struggling with. An additional challenge is that the user expects at-least the phones to work
during an outage. This is hard to accomplish with VOIP technology which we are expecting to deploy. While the
staff was being trained, the workstation power cables started to melt from the excessive voltages and voltage
fluctuations. We have learned the hard way that power cables from the US are rated only at 130 VAC and perform
poorly at the higher 230 VAC continuous loads.
*  Backups The application is backed up on an aggressive schedule using automysqlbackup. The file server
is backed up using Bacula and the firewall is backed up via a shell script.
*  Please Contact me if you are interested in this project at sidhu@medigrail.com