[EpiData-list] outbreak table and calculations

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Wed Aug 11 12:53:08 CDT 2010


Thanks for the suggestions on creating condensed tables.

 

As you know, I am trying to learn how to use EpiData Analysis by
practicing with a previous outbreak that occurred. However, I'm running
into some troubles with the calculations. I wonder if I'm missing a
command or doing something wrong.

 

For example:

 

I have 65 records. 13 cases. 51 controls. 1 excluded.

 

When I type:

 

Freq item03c 

 

Item03c

 

N

Y

18

N

27

Total

45

 

When I type:

 

Tab status item03c

 

Status

Item03c

Case

Control

Total

Y

9

9

18

N

3

24

27

Total

12

33

45

 

So far, this matches with the data in my .dbf file. I manually counted
exposed/unexposed & case/control for item03c.

 

When I type:

 

Tab status item03c/oa

 

Outcome: status by item03c

Status by item03c

Exposed

Not Exposed

 

 

 

N

n

Ill

AR(%)

N

n

Ill

AR(%)

RR

(95% CI)

 

45

27

24

88.9

18

9

9

50

1.78

(1.10-2.88)

Exposure: (Item03C=N)

Outcome: Status=control

 

I think there is something wrong with these calculations. It seems that
the wrong numbers are being used to calculate the AR and RR. Why in this
table does it have exposed n=27 instead of exposed n=18 and exposed
Ill=24 instead of exposed Ill=9 and vice versa with the not exposed
values? When I manually calculated these numbers in a 2x2 table, I
assumed that the table would look like this and the calculations are
these:

 

Status

 

Case

Control

Total

AR(%)

RR

Exposed

9

9

18

50

4.5

Not exposed

3

24

27

11

 

Total

12

33

45

 

 

 

What am I doing wrong? Do you have suggestions for my problem?

 

Thanks,

Sandy

 

--------------------------------

Sandy Goeke, MPH

sandy.goeke at co.benton.or.us

 

-----Original Message-----
From: epidata-list-bounces at lists.umanitoba.ca
[mailto:epidata-list-bounces at lists.umanitoba.ca] On Behalf Of
epidata-list at lists.umanitoba.ca
Sent: Monday, August 09, 2010 11:46 PM
To: epidata-list at lists.umanitoba.ca
Subject: Re: [EpiData-list] How to define Jumps at a variableusing
anothervariable

 

  Three ways:

a. use condensed table format:

    tab casevar exp1 exp2 exp3 ......exp50        /oa

 

b. Open a logfile, e.g.

    logopen outbreak.htm

    tab casevar exp1 exp2 exp3 ......expn        /oa

    logclose

    * Then you can view the file outbreak.htm in your standard browser.

 

c. If you prefer single tables rather than the condensed one:

    Select 25 variables at a time

 

 

regards

 

Jens Lauritsen

EpiData Association

 

> I am trying to analyze data for an outbreak. I am trying to run 2x2

> tables for 50 food items. After running about 40 tables, the results

> viewer window refreshes and then only shows the later 10 tables. I am

> unable to scroll up to find the results for the first 40 tables. How
do

> I retrieve that information? How do I view the results for all 50 of
the

> tables?

> 

> Thanks in advance for your input.

> 

> Sandy

> --------------------------------

> Sandy Goeke, MPH

> sandy.goeke at co.benton.or.us

> 

> 

> _______________________________________________

> EpiData-list mailing list

> EpiData-list at lists.umanitoba.ca

> http://lists.umanitoba.ca/mailman/listinfo/epidata-list

> 

_______________________________________________

EpiData-list mailing list

EpiData-list at lists.umanitoba.ca

http://lists.umanitoba.ca/mailman/listinfo/epidata-list



More information about the EpiData-list mailing list