Print

Print
Thank you
Do studies report  how they arrive at patient years

Amy

Sent from my iPhone

On Oct 26, 2012, at 4:25 PM, "Juan Acuna" <[log in to unmask]> wrote:

... It means that there were a number of outcomes in100 years of followup (from a potential endless combinations of time span of followup per person). It is a denominator in cohort data where individuals contribute different times of followup until the given outcome presents. it certainly does not mean "a 100 persons followed for a year". that would be worded something like "outcome per 100 persons per year" , such as pregnancy rates for contraception were reporting is "5 pregnancies per 100 women per year" (Although some use the person-year denominator as well).

In the study, 105 deaths per 100 person-year means 105 deaths in a 100 [combined] years of followup...

Juan M. Acuna M.D., MSc., FACOG.

Associate Professor of Human and Molecular Genetics and Clinical Epidemiology, FIU Herbert Wertheim College of Medicine

FIU Assistant Vice President for Clinical and Community Research

Director, Division of Research FIU HWCOM

Guest Researcher,
Centers for Disease Control and Prevention WHO-PAHO  
Collaborating Center

11200 SW 8th Street
AHC2-474
Miami, FL 33199

Ph (305) 348-0676


On Oct 26, 2012, at 4:04 PM, "Ashwin Singh" <[log in to unmask]> wrote:

Hello,

This is not exactly an answer to this post but more of a query i have.

Does 100 person-years necessarily mean 100 persons followed for 1 year? Or it may also mean 10 persons followed for 10 years or a single person followed for 100 years?

Im not sure about the concept of dying before reaching one year. It cant work out that way. Because, one would have to accomodate all people dying by the end of one year and not individually as to how soon before the end of one year in the present equation.

Would be great to have a better understanding on this.

Best!

On Fri, Oct 26, 2012 at 11:22 PM, Stephanie Chan <[log in to unmask]> wrote:
Hello all,

I'm reviewing the BMJ validation article on the CHADS-VaSC score for predicting thromboembolism in afib, and I'm trying to figure out how to best interpret the term "rate per person-years."  I understand that when a rate is list as "5 deaths per 100 person-years," that means that if 100 people are followed for 1 year, 5 will die.  Is it fair to say that that's the same as a 5% annual death rate (assuming a constant rate of death in the subsequent years)?  

One reason I'm asking is that I noticed that the death rate of those with a CHADS-VaSC score of 9 has a mortality rate of 105 per 100 person-years -- but you can't have a 105% mortality rate!  I assume that this is because a substantial number of people are dying before they reach 1 year, so the denominator is less than one.  Is this correct?

Thanks for any help you could offer!

Stephanie Chan, M.D.
Editor-in-chief







--
 
Best regards,
 
Ashwin Singh Parihar
M.B.B.S Final year,
V.C.S.G. Govt. Medical Science & Research Instt. Srinagar,Garhwal
Uttarakhand.