USMLE Forums banner
1 - 7 of 7 Posts

·
Registered
Joined
·
112 Posts
Discussion Starter · #1 ·
I am confused about how to calculate these 2 parameters, for example if we have 100 new cases, total population 1000. Some books say incidence should be number of the new cases divided by the number of the total population AT RISK. It means 100/ 1000-100 which 100/900 because they consider only 900 are at risk as the 100 new cases already have the disease. Other books just divide 100 by 1000. In first aid they put that comment about incidence only (When calculating incidence, don't forget that people currently with the disease, or those previously positive for it, are not considered at risk).

So which method should we use for the exam and does it apply for incidence and not for prevalence ?
 

·
Registered
Joined
·
4,342 Posts
It's the time you look at the population that decides

It depends on when you look at the population.

There are two types of Incidence Rates:

Incidence Rate: New cases (events) / Population at risk over a period of time, in your example 100/900

Cumulative Incidence Rate: New cases (events) / Population at risk at the beginning of the study, in your example 100/1000

The same applies to prevalence, always the denominator is the population at risk, it's just the time you look at them.
 

·
Registered
Joined
·
40 Posts
Would it be possible to clarify on this question, in regard to prevalence?

First Aid refers to both denominators (of incidence and prevalence) as the population "at risk." Because of this, in a population of 100,000, of which 1,000 are HIV positive, is the point prevalence of HIV(+) individuals:

(1,000 / 100,000), or
(1,000 / 99,000)?

In a population of 100,000 (without death / migration / etc), of which 1,000 subjects are already HIV(+), and each year an additional 100 patients become infected with HIV, is the period prevalence over 5 years equal to:

(1,500 / 100,000),
(1,500 / 99,000), or
(1,500 / 98,500)?

Thanks for your time and help,
Tig
 

·
Registered
Joined
·
1,357 Posts
Edit:
Cumulative Incidence is the same as just Incidence = at this time, how many cases we have = 100/1000...numerator is just a subgroup of the denominator.

An incidence rate, or simply rate, is defined in terms of number of new cases per observed person-time (rate)
 

·
Registered
Joined
·
40 Posts
IR = period time in this case 1 year = 100/1000 (entire population was at risk in that particular period of time)

Cumlative Incidence is the same as just Incidence = at this time, how many cases we have = 100/1000...in this case both are the same
Could you define the acronyms you use (IR)?

Also, my question was regarding point-prevalence and period-prevalence, not incidence.

Thanks,
Tig

(edit: that is, unless you're responding to the OP instead of me)
 

·
Registered
Joined
·
1,357 Posts
Could you define the acronyms you use (IR)?

Also, my question was regarding point-prevalence and period-prevalence, not incidence.

Thanks,
Tig

(edit: that is, unless you're responding to the OP instead of me)
The point prevalence (or prevalence proportion) is the proportion of people who have the disease at the time of the study and
The period prevalence is the proportion who have had the disease over some past time interval (such as the past 2 years)

and btw, if they ask for IR...we should use: new cases per observed person-time (rate)...;)
 

·
Registered
Joined
·
113 Posts
It depends on when you look at the population.

There are two types of Incidence Rates:

Incidence Rate: New cases (events) / Population at risk over a period of time, in your example 100/900

Cumulative Incidence Rate: New cases (events) / Population at risk at the beginning of the study, in your example 100/1000

The same applies to prevalence, always the denominator is the population at risk, it's just the time you look at them.
There is only ONE type of Incidence "Rate"…Cumulative Incidence is NOT a RATE!…it´s just a proportion or a ratio.

Incidence (aka Cumulative Incidence) = # NEW cases in specified period / # people at risk AT the BEGINNING of the specified period

You MUST include the NEW cases in the DENOMINATOR!!!

In INCIDENCE Rate (a "REAL" RATE!), the DENOMINATOR is expressed in "person-time" units (person months, years, etc.)

Go to this web site (tufts university) for more examples
http://ocw.tufts.edu/Content/1/lecturenotes/194069/194120
 
1 - 7 of 7 Posts
Top