November 2008 - EpiData-list - University of Manitoba Mailing Lists

Dealing with 'Ongoing' duration in epicurves
by epidata-list＠lists.umanitoba.ca 02 Dec '08

02 Dec '08

I would appreciate it if those working with outbreaks could reply about how they enter and display "ongoing" duration of illness. I have been given an Excel sheet to analyse in Epidata. For this outbreak people were asked the date and time when they became unwell and the date and time when they felt better. There is a calculated field in the spreadsheet that calculates duration in days (date better - date unwell), but for some people this is empty because "ongoing" has been entered as text in the "date better" field. I want to modify the Excel sheet so I include those people with 'ongoing' illness in an epidemic curve. I thought I could enter a date sometime in the future instead of 'ongoing' This would mean the duration days would not be blank, but a number of days well beyond the maximum value for those whose illness is finished. What do others do? thanks Christine Christine Roseveare Analyst Regional Public Health 570 9194 027 495 9671 -- The information contained in this email and any attachments is confidential and may be legally privileged. If you have received this message in error, please notify the sender immediately and remove all copies of the message, including any attachments. Any views or opinions expressed in this email (unless otherwise stated) may not represent those of Hutt Valley DHB. Thank you. ** Disclaimer added by HVDHB **

1 2

Export to SPSS requires 2 stage procedure
by epidata-list＠lists.umanitoba.ca 30 Nov '08

30 Nov '08

Hello users and developers, I am using Epidata Entry 3.1 27 Jan 2008 on Windows XP Pro SP3. I have date entered in .REC files. When I perform the export function using the drop down menus, it does so fine, but the .SPS and .TXT files created are all blank. If I try re-exporting (with Epidata Entry still open), and error message pops up: "...data still in use..." or something like that. A successful data export to SPSS is only possible if I complete quite Epidata Entry and and start a new instance to perform the export procedure once again. Could somone please verify this for me, and if so please explain what is going on. It is quite frustrating that to have to do export in 2 stages. Thanks, Shuaib

1 0

Re: Test of a new "set" principle
by epidata-list＠lists.umanitoba.ca 27 Nov '08

27 Nov '08

The very quick message from N. Verlander was the observation that: set option means = ? option means not supported . set option means = "/t" OPTION MEANS = "/t" set option tables = ? tables sex kmgrp option tables = /t /r tables sex kmgrp /t /r (but the user wrote the same as above) The message of the default options NOT defined is correct. But we could/should most likely change the message from: option table not supported to option table not defined Another issue was that the option /t2 was not defined for Ichart. This will add two tests: test 4: 2 out 3 successive points more than 2 standard deviations (sigma) away from the centerline. test 5: 4 out of 5 successive points more than one standard deviation (sigma) away from the centerline. and currently also lines at 1 and 2 sigma from the centerline, but only for Ichart. A detailed document on respecification of spc charts is now available on the testing page. -- Jens Lauritsen Coordinator and initiator of EpiData Project http://www.epidata.dk

1 0

Test of a new "set" principle and question on definition of SPC commands
by epidata-list＠lists.umanitoba.ca 27 Nov '08

27 Nov '08

At the test page http://www.epidata.dk/testing.php a new build (158) of EpiData Analysis has been placed for testing and commenting here on the list. The main new aspect is a change in user specifications for set commands as a consequence of the discussion on this list last week. For all commands each user can define his/her own options, e.g. if you desire row percentage and chi square tests always for tables add to your epidatastat.ini file: set option tables = "/t /r" If you have the custom instead to use the short form of tables you would write: set option tab = "/t /r" Two "command group" set commands are predefined: set OPTION GRAPH (all graphs excluding SPC graphs) set OPTION SPC (SPC charts including pareto chart and run chart) Another area for testing is the extended SPC definitions, with new graphs: ichart /mr, gchart, cchart,uchart and ichart /t2 A suggestion has been made to change SPC charts such that the "time" variable is only used when the user specifically wishes to do so. It is expected that after a certain build/release the syntax for SPC charts will be: Xbar-S measurement‭ ‬time Ichart measurement‭ [‬time‭/label]* Uchart count volume ‭[‬time/label‭]* Cchart count‭ [‬time‭/label]* Pchart count total‭ [‬time‭/label]* Gchart "time at‭ ‬incident‭"‬ (only one variable) Pareto categories‭ Where [time/label]* indicates that this variable is NOT mandatory. If given this will be used as a label (like /xlabel), when the variable is string and as a value for x when it is numerical. Comments to this change are welcomed. -- Jens Lauritsen Coordinator and initiator of EpiData Project http://www.epidata.dk

1 0

Documentation of lifetable updated
by epidata-list＠lists.umanitoba.ca 27 Nov '08

27 Nov '08

Branger pointed out that the documentation on lifetable command in latest analysis was faulty. An updated version of commands.htm has been uploaded and can be found at: http://www.epidata.dk/commands.htm#lifetabgrp The file will be added to install versions later. Estimation options: /time=x :Calculate Survival estimate at indicated time value (values) /p25 p50 p75 :Calculate Time span estimates at Quartiles and Median of Survival Proportions (0.75 0.5 0.25) /t : Log-rank test and Hazard Ratio (only relevant with /by) /ref=x : Indicates value of /by variable to use as reference for comparisons when the difference from reference is estimated for /time /p and /t options The option /G has been replaced with /NG to hide the graph., but was incorrect in the documentation. Hide/show: * /NG do notshow the Kaplan Meier plot * /NT : do not show summary documentation table * /NOLT: do not show lifetable * /E0 /E1 /E2 /E3 /E4 : show results with this number of decimals * /NOCI : Hide confidence intervals in tables and graph (e.g. if graphs overlap for groups) regards Jens Lauritsen EpiData Association

1 0

Epidata
by epidata-list＠lists.umanitoba.ca 26 Nov '08

26 Nov '08

Hello, I have a problem with relate for two related files (with merge command), and its OK with EPI INFO 6 (DOS) The key is fiche. I wrote these commands (I simplify) : read first.rec merge fiche /file=second.rec and is written : Non-unique key in: second.rec The two files .chk are : - first.rec FICHE KEY UNIQUE END ONEVAR IF ONEVAR=1 THEN RELATE FICHE SECOND.REC ENDIF END - second.rec FICHE KEY 1 END Thank you PS : In lifetable Help /haz and /G are notified but these commands are not effective Bernard BRANGER - Médecin coordinateur Réseau de Santé en Périnatalité "Sécurité Naissance - Naître ensemble" des Pays de la Loire 1, allée Baco - 44000 NANTES Tél 02 40 48 55 81 - Fax 02 40 12 40 72

1 1

Numeric Range for field (two ranges)
by epidata-list＠lists.umanitoba.ca 26 Nov '08

26 Nov '08

1. Numeric Range for field (epidata-list(a)lists.umanitoba.ca) > Hello, > In the .chk file, I want to define a range for a numeric field (v1 ###) as follows: 100-120, 200-250 > Easiest solution: v1 range 100 250 after entry if (v1 > 120) and (v1 < 200) then help "Range allowed: 100-120 or 200-250" goto v1 endif end end Regarding the other question yesterday on loss of date we need to know more. regards Jens Lauritsen EpiData Association

1 0

Changing id value
by epidata-list＠lists.umanitoba.ca 24 Nov '08

24 Nov '08

Can anyone tell me how to change the value of in my id? When ever I change the value some of my data disappears. Thanks Raphael

1 0

Numeric Range for field
by epidata-list＠lists.umanitoba.ca 24 Nov '08

24 Nov '08

Hello, In the .chk file, I want to define a range for a numeric field (v1 ###) as follows: 100-120, 200-250 (Break the continuity) But I'm trying generates an error. Note: Epidata accept: 100-120, 200 In addition tried v1 RANGE 100 120 RANGE 200 250 END Then, open the REC but when entering data and a message "is not a valid floating point value" What's happening? and how can I set the range for my values, without defining them individually? Cordially Carlos Ortega Areyte Biostatistics from Chile, South America -- Carlos A. Ortega Areyte Prof. de Matemática y Computación Magíster en Bioestadística ACADÉMICO

1 0

Select, if and count giving different numbers in EpiData Analysis v2x
by epidata-list＠lists.umanitoba.ca 21 Nov '08

21 Nov '08

For certain logical statements the following commands would work on different subsets of the data read myfile sort samlet select if (samlet = samlet[_n-1]) count if (samlet = samlet[_n-1]) test = test + "x" if (samlet = samlet[_n-1]) if (samlet = samlet[_n-1]) then test = test + "x" .................................................................................................................... The bad news is that this is very bad behaviour. The good news is that the cause of the error has been located and a new build (v2.1.0.157) has been uploaded. If you are not working with selects and if statements this will not affect your work. Strange that no-one found the bug before (I found this yesterday) and the test environment has been extended with this aspect. regards Jens Lauritsen EpiData Association

1 0