what is difference between Rate ratio and Risk ratio?

[thenmozhi mani]

Hi

what is difference between Rate ratio and Risk ratio?
how to estimated prevalence rate ratio or risk ratio in observational study?

--
Thenmozhi

[Martin Holt]

Hi Thenmozhi,

 

There seem to be a lot of questions about definitions right now !

 

Rate ratio: ratio of the rate of an event in one group (exposure or intervention) to that in another group (control).

 

Risk ratio: ratio of the risk of an event in one group (exposure or intervention) to that in another group (control).

 

So it depends on your definitions of rate and risk.

Rate:summary measure which conveys the idea of risk over time. Calculated as number_of_new_occurrences_of_a_particular_event/person_time_at_risk

 

Risk: probability of occurrence of a given event. Calculated as number_of_events/number_of_people_at_risk

 

The term "relative risk" is sometimes used as a synonym for risk ratio, and rate ratio is one of the relative risk measures too.

 

Hope that helps,

 

Martin

 

Ref."A-Z of Medical Statistics" by Filomena Pereira-Maxwell

[thenmozhi mani]

Hi Martin,

 

What do u mean... Rate ratio and risk ratio are same or different. Are incidence and prevalence anything to do with these measures.

--
Thenmozhi

[Adrian Sayers]

i recommend sander greenland modern epidemiology, as everything it
depends on defn!

Good luck

Adrian

2008/10/14 thenmozhi mani <mani.th...@gmail.com>:

[Martin Holt]

Hi Thenmozhi,

 

I'm sure you're just as capable of looking these things up as I am (:.

 

Relative risk is a ratio of the probability of the event occurring in the exposed group versus a non-exposed group.

Another term for the relative risk is the risk ratio because it is the ratio of the risk in the exposed divided by the risk in the unexposed.

Rate ratio is not so straightforward,see http://www.jstor.org/pss/3702753 . I've seen rate ratio called "incidence rate ratio".

You might find http://www.bmj.com/cgi/content/full/333/7581/1248 interesting.

My conclusion would be that risk ratio is not the same thing as rate ratio, but that under certain conditions they are not so far apart. Other people may choose to differ - as Adrian says, it depends on the definitions.

I recommend that you try researching these terms yourself - you will experience the lack of black-and-white clarity for yourself. But do come back to the group if you need to.

Best Wishes,

Martin Holt

 

[BXC (Bendix Carstensen)]

I think we could be a bit more precise here:

A rate-ratio is a ratio of rates, a risk ratio a ratio of risks.

A rate is a parameter in a model:

rate(t) = lim(dt->0) P( event in (t+dt) | at risk at t ) / dt

Under the assumption of the rate being constant, i.e. rate(t) = \lambda, say, the maximum likelihood estimate of lambda, based on D event dyring Y time of follw-up, is

\hat \lambda = D/Y

Note that this is a scaled quantity that is measured in units of time^(-1).

A risk is a probability of an event over a prespecified interval, (a,b) say:

risk(a,b) = 1 - exp( -\int_a^b rate(u) du )

(this is conditional on being alive at a).

The maximun likelihood estimator of the risk is just the corresponding transformation of the mle of the rate. Note that this is an extremely arrogant statement as is only is feasible to compute if the rate is assumed constant, in which case it is

1 - exp( -(b-a)D/Y )

And you will note that the argument to exp is dimensionless since Y and (b-a) both have dimension time.

You can find a more extensive account of the relationships between rates etc. (in a readable typography too) on pages 5-7 of:
http://staff.pubhealth.ku.dk/~bxc/AdvCoh/Aalb-2008/pracs.pdf

Best regards,
Bendix
______________________________________________

Bendix Carstensen
Senior Statistician
Steno Diabetes Center
Niels Steensens Vej 2-4
DK-2820 Gentofte
Denmark
+45 44 43 87 38 (direct)
+45 30 75 87 38 (mobile)
b...@steno.dk http://www.biostat.ku.dk/~bxc

> -----Original Message-----
> From: MedS...@googlegroups.com
> [mailto:MedS...@googlegroups.com] On Behalf Of Martin Holt
> Sent: 14. oktober 2008 11:30
> To: MedS...@googlegroups.com
> Subject: {MEDSTATS} Re: what is difference between Rate ratio
> and Risk ratio?
>

[BXC (Bendix Carstensen)]

[Thenmozhi Mani]

Dear all,

Thanks to all of you.....

--
Thenmozhi