Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Calculating duration of time

From   "Schonlau, Matthias" <[email protected]>
To   "'[email protected]'" <[email protected]>
Subject   st: RE: Calculating duration of time
Date   Wed, 11 Dec 2002 17:50:17 -0800

This makes sense if time is a  discrete variable 
  e.g. 5th-10th day  where 5 refers to a full day and 10 refers to a full
  total time= 6 days
It doesn't make sense if time is continuous.
 e.g. 5th day midnight  -10th day midnight  
  total time = 5 days. 


-----Original Message-----
From: Joseph Coveney [mailto:[email protected]]
Sent: Wednesday, December 11, 2002 5:08 PM
To: Statalist
Subject: st: Calculating duration of time

Somewhere I had read that it has become a convention to define a duration or
of time as:

end time - start time + 1,

where the 1 is the unit of time used.  Thus, for example, for a common
economics measure, the length of hospital stay (LOS) is defined to be
discharge date - 
admission date + 1 day.

A patient who is discharged a few hours after admission would have an LOS of
day, one who is discharged the day after admission would have an LOS of two
and so on.  I suppose that the motivation for this definition is to remove
zeroes from 
durations, since these are often skewed and a logarithmic transformation is
used, as for 
example -means LOS, add(1)-.

I thought that I had seen the formula in an online data definition document
or data 
standards document at a website for the Center for Disease Control, FDA, HL7
CDISC, but I cannot relocate it.  Apologies for the somewhat off-topic
nature of the 
question, but has anyone on the list come across this convention?

Joseph Coveney
*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index