See AK> Below. Sounds great! -------------------- Andrew S. Kanter, MD MPH
- Director of Health Information Systems/Medical Informatics Millennium Villages Project, Earth Institute, Columbia University - Asst. Prof. of Clinical Biomedical Informatics and Clinical Epidemiology Columbia University Email: [email protected] Mobile: +1 (646) 469-2421 Office: +1 (212) 305-4842 Skype: akanter-ippnw Yahoo: andy_kanter >________________________________ >From: Jose Garcia Muñoz <[email protected]> >To: [email protected] >Sent: Saturday, September 3, 2011 9:21 AM >Subject: [OPENMRS-IMPLEMENTERS] Starting a Project based on OpenMRS >scalability in isolated areas of Perú (EHAS Foundation) > > > >Dear all, > >EHAS Foundation from Spain (www.ehas.org) are interested in using OpenMRS as a >Health Information System centered in the Patient Medical Record in Perú to >improve the quality of the clinic information and the decision making, and reduce the time waiting for a medical assistance. More specifically, we are working with the local government of Loreto (it's a Peru region bigger than spain situated in the middle of the Amazon Jungle, where the health facilities are around 300), just to make an analysis of the clinical and epidemiological information they need. > > >AK> Excellent, there are lots of resources you can get help with. However, who >is going to be doing implementation and support? How many people do you have >and how are you going to customize and maintain the code/content? Do you have >developers? > >At the moment we are centering our work in primary health care, and just to >reduce the complexity of the problem we are focus on the most important >clinical data they use. So, we have to collect clinical information focusing >on patients associated with malaria, diarrheal diseases and acute respiratory >diseases, although also it's important to collect some (minimal) information >related to other diagnosis and pathologies. To be more specific: > >1) we work with DIRESA (local government of Loreto) and 14 health facilities related to primary health care. These facilities are connected by a WiFI local network covering 500 km (more or less. Each health facility is located 50 km from another). > > >AK> Connectivity is a major issue. Have you tested that you can use OpenMRS >with workstations in one place and the server in another? Do you have 24/7 >access? > >2) These health facilities are divided in 2 health centers and 12 health post. Each health center is the head of a health micro-network (it consists of a set of health post and one health center), so all the information (and referral patients) go from the Health posts to Health center and finally arrives to DIRESA. > > >AK> We have similar arrangements in MVP. We currently bring in health post >data to be entered into OpenMRS at the central clinic. However, with the use >of Android platforms it is possible to use phones in the health posts. >However, this depends on connectivity, staffing, etc. > >3) we will collect a limited set of clinical information. For the health posts is enough with patient demographic data, ICD-10 diagnosis, some dates and questions, and maybe some vital signals (temperature, respiratory, ...) (anyway diagnosis and dates related to are the most important data). Health centers will collect a bit more sophisticated information specially related to lab test of malaria, diarrheal diseases and acute respiratory diseases. > >AK> Sounds like a good focus. >4) >Clinical information generates different reports that are sending (weekly or monthly) from health posts to health centers, and from health centers to DIRESA. Of course we want to generate this reports automatically for three reasons: (a) Reduce the time of workers filling paper forms, (b) Reduce the time of workers traveling with the paper forms to the health centers or DIRESA (it takes two days per trip), and (c) Increase the quality of information. >AK> What are you doing for reporting? I think PIH would have a lot to offer >here, but eventual integration with a DHIS system like DHIS2 would be best for >automated district health information system reporting. > >In terms of clinical paper forms and data to collect we have enough >information, but, >i think there are some important tasks or goals to pay special attention. Marta (in copy in this email) is trying to analyze the scope of OpenMRS in terms of scalability and interoperability with other openMRS servers as well (also for future the idea is just to implement OpenMRS at every centers we can). Cause we don't know the whole scope of the system (what it's implemented, designed,... or not), we'll appreciate any advice, experience or support you can provide for that. Again, specific tasks: > >1) OpenMRS deploy: For now, maybe one OpenMRS server located in the DIRESA's offices will be enough. but looking ahead we need something more flexible. In future we will have thousands of medical registers (much information for only one server i think). We need something very flexible in terms of hierarchy and Sync with other OpenMRS system. This month we start working on the analysis of 'Remote Form Entry Module' and 'Sync Module' modules. Who has experience rolling out OpenMRS that way?. We really appreciate some tips. suggestions, or email exchange about using OpenMRS as a distributed system. > > >AK> Again, PIH can chime in here. There are some interesting experiments using >CouchDB to do replication, but your decision will probably depend on how >reliable your connectivity is and whether you need a local store or can do >everything in thin client mode. > >2) ICD-10: It's very important for us to store all the diagnosis in ICD-10 code. For now, we have identified the most important ones in Loreto Region (which are used for the epidemiological reports). They are around 40 different diagnosis, so there is no problem to introduce them manually in the concept dictionary. Again, for next steps, sure we'll need to introduce more sets of ICD10 codes and they are around 13.000. Do you know about the existence of some script (or module) that can provide the dump of that amount of information? OpenMRS Concept Cooperative (OOC) could help? > > >AK> The MVP/CIEL dictionary is ICD-10 coded as well as SNOMED-CT coded. We >don't load in ICD-10 as descriptors as they are not clinically friendly. If >you have any problem finding the ICD-10 coded concept you can use >maternalconceptlab.com search to search by ICD-10 code. If you still can't >find it, let me know and I will be sure that it is included. > >3) It's very important to associate each patient encounter with the location (also with its geographic position) and with the doctor. The main reasons: > > * Clinical responsibility: The Peru's government demands to know the > responsible of the medical care, > > * Control access: A doctor of a health center can access to the medical > records of patients in both the health center and the health posts associated > to it. A doctor of a health post only can access to the patients treated in > his facility. > * Custom forms: Health center forms will be more sophisticated (it will contain more clinical information) than the health posts ones. > * Analysis of data and measuring the quality information: One of our final main goals is to study the connection between the clinical information, its geographic location, and the flows of this information. we are fairly sure that data mining techniques focusing on this information can extract indications of disease patterns, health care weakness (comparison between regions could help identify most cost effective improvement strategies), population weaknesses, help formulate clinical strategies for each region, etc.. >AK> Not an issue. You can consider having one master location table or use >something like address hierarchy to ensure patient location data is properly >recorded. You should distinguish between residence and location of clinical >care. You can also use OpenMRS to record GPS coordinates. AMPATH has been >doing some of this with their community based care module. > >Thank you very much. > >Best regards >Jose >>________________________________ > Click here to unsubscribefrom OpenMRS Implementers' mailing list > > _________________________________________ To unsubscribe from OpenMRS Implementers' mailing list, send an e-mail to [email protected] with "SIGNOFF openmrs-implement-l" in the body (not the subject) of your e-mail. [mailto:[email protected]?body=SIGNOFF%20openmrs-implement-l]
