Datetime conversion resource
Hi list members From time to time on the list we have discussed the value of having a datetime variable in Epidata. This is something I would find like, as the 'epidemic curve' is something that is an important part of any outbreak investigation. For short outbreaks that carry on over more than one day both the time and date are useful so youcan show numbers becoming ill every 6 hours for example over two or three days. The ability to map by datetime is one reason I often use EpiInfo for outbreak analysis, although I enter data in Epidata. I wanted to let any interested members know that a colleague has developed a simple text file that can be run as a programme in EpiINfo that coverts data entered in EPidata to a single datetime for use in EPiInfo. If you have entered an onset date as a date and onset time as a number with a decimal (e.g. 15.30) the programme converts the information into a single datetime variable - it can then be mapped using the EpiInfo histogram graph. Jens if you think this is useful at all would you like me to post this somewhere? Also is there any thought of creating a time field in Epidata?
thanks
Christine
Christine Roseveare Analyst Regional Public Health 570 9194 027 495 9671
epidata-list@lists.umanitoba.ca 22/10/2010 10:47 p.m. >>>
For all users
Today new versions of Entry Client and Manager have been placed on thehttp://epidata.dk/testing.php page for general download and testing.
Apart from some minor points and errors reported to the Flyspray bug database the current release is what we see as the first release candidate for general usage following key developments mentioned below.
level 1: Basic functionality at field level level 2: Extended functionality with value labels, basic checks etc
We are at level 1 now and production release will include level 2.
Main news since last release are: a. bug fixes and enhancements b. availability of two file formats: epx: the basic xml structure containing all information for a given project epz:The same file, but zipped, which often reduces the overall size to about 10-30% of the epx file c. export of data to stata d. short introduction to the EntryClient
Currently the users should find and test the following: Manager a. all aspects of creating new projects with one database table/file per project a1. importing (loading) old rec files and stata files a2. Pasting from qes files or editor files onto a dataform a3. Saving these data as a project file with options for entry b. exporting data to Stata (regardless of whether you have Stata)
EntryClient b. data entry of project files created with manager b1. Securing that data are saved and shown as read b2. Editing and browsing in data b3. Backup of files after entry b4. Timed recovery function if systems break down during entry
What are the next steps ? c1. Adaptation of EpiData Analysis to be able to read the xml structured files (epz and epx) c2. Following users tests in the coming weeks further error removal will be done in EntryClient and Manager Please report if possible to the flyspray database system. (But also if you cannot get the flyspray system to work let us know on this list)
Kind regards
Torsten Christiansen, Jens Lauritsen EpiData Association
As you might have noticed in the new manager /entryclient sample file there is a time field. In other words this means that once we convert to the new EPX file format there is a time field (hh:mm:ss).
When you enter data in this type of field, two digits will be hour, four will be hour and minute and six will include seconds. Parts not filled in (less than six digits) are automatically saved as 00.
My experience from my daily work in the hospital here and reporting to national clinical databases it is quite annoying to enter data as a "true" date-time field, because one (here) has to use the mouse and click for entering date.
What is planned therefore is NOT a date-time field, but functions to combine a date field with a time field for export.
And obviously for graphs like Epicurves the x-axis will be able to use a data+timefield to use for graphing.
Regards Jens Lauritsen
Den 26-10-2010 00:49, epidata-list@lists.umanitoba.ca skrev:
Hi list members From time to time on the list we have discussed the value of having a datetime variable in Epidata. This is something I would find like, as the 'epidemic curve' is something that is an important part of any outbreak investigation. For short outbreaks that carry on over more than one day both the time and date are useful so youcan show numbers becoming ill every 6 hours for example over two or three days. The ability to map by datetime is one reason I often use EpiInfo for outbreak analysis, although I enter data in Epidata. I wanted to let any interested members know that a colleague has developed a simple text file that can be run as a programme in EpiINfo that coverts data entered in EPidata to a single datetime for use in EPiInfo. If you have entered an onset date as a date and onset time as a number with a decimal (e.g. 15.30) the programme converts the information into a single datetime variable - it can then be mapped using the EpiInfo histogram graph. Jens if you think this is useful at all would you like me to post this somewhere? Also is there any thought of creating a time field in Epidata?
thanks
Christine
participants (1)
-
epidata-list＠lists.umanitoba.ca