cumulative vs incience rate

[aioo]

Hi,

I am always confused about the difference between incidence rate and cumulative incidence. I know the defenition on both but find it hard to recognize in real articles.

Example:
“The Kaplan-Meier event rate fort he primary end point at 7 years was 32.7% in the simvastatin-ezetimibe group, as compared with 34./7% in the simvastatin-monotherapy group (absolute riskdifference, 2.0 percentage points; hazard ratio, 0.936; 95% confidence interval, 0.89 to 0.99;P=0.016).

are the percentages incidence rate or cumulative incidence and why?

2nd example:

“Of 477 patients, 139 developed recurrent VTE over the course of 1533 person-years of follow-up. The adjusted 10-year VTE recurrence rate was 28.6%. The adjusted 90-daymajor bleeding on anticoagulation rate was 1.9%.”

---

Members don't see this ad.

 

[Winged Scapula]

Unfortunately, your links are in Dutch, so the vast majority of users here aren't able to read them

 

[aioo]

Unfortunately, your links are in Dutch, so the vast majority of users here aren't able to read them

I am sorry for that. I have removed them now. But they weren´t relevant for the question anyways. We get just those tekst that are written, and then are expected to tell if those are cumulative incidence or incidence rate.

 

[PsyDr]

Hi,

I am always confused about the difference between incidence rate and cumulative incidence. I know the defenition on both but find it hard to recognize in real articles.

Example:
“The Kaplan-Meier event rate fort he primary end point at 7 years was 32.7% in the simvastatin-ezetimibe group, as compared with 34./7% in the simvastatin-monotherapy group (absolute riskdifference, 2.0 percentage points; hazard ratio, 0.936; 95% confidence interval, 0.89 to 0.99;P=0.016).

are the percentages incidence rate or cumulative incidence and why?

2nd example:

“Of 477 patients, 139 developed recurrent VTE over the course of 1533 person-years of follow-up. The adjusted 10-year VTE recurrence rate was 28.6%. The adjusted 90-daymajor bleeding on anticoagulation rate was 1.9%.”

You seem to be conflating terms. Incidence is used in a different process or area of interest.

Event rate is best conceptualized as the percent in one group. In your first example there are two groups. The metric described is the percentage for group a and the percentage in group b. Not cumulative.

 

[aioo]

You seem to be conflating terms. Incidence is used in a different process or area of interest.

Event rate is best conceptualized as the percent in one group. In your first example there are two groups. The metric described is the percentage for group a and the percentage in group b. Not cumulative.

thanks for the response. But I am sure that in both examples the right answer is cumulative incidence. This comes from mine medical school exam.

In the first example: it is true that there are 2 groups but they are asking to describe what kind of rate measure they have used in that example, and the correct answer is cumulative incidence. So i am trying to understand why that is the case and how cumulative incidence can be differentiated from incidence rate when you read a piece of text like that.

Here is an other example from my exam:

“The risk of celiac disease at 5 years of age was 12.1% (95% CI, 9.2 to 15.0).”

and the right answer here is also cumulative incidence. My question remains the same: how can I differentiate between these two forms of incidence when i read a text like this. Thanks

 

[AcronymAllergy]

Quick answer, assuming I'm remembering/explaining my epidemiology correctly:

Cumulative incidence is the number of people developing a condition over a period of time. So in the first example, 32.7% of patients in group A and 34.7% of patients in group B developed the condition, symptom, outcome, etc. of interest over the specified time period (which looks to be 7 years). It's actually more of a proportion than a true rate. That is, it's the proportion of folks who developed the condition of interest over a specified period of time. The numerator is the number of cases and the denominator is the number of people followed. Cumulative incidence must always specify a period of time.

Incidence (or incidence rate) is the rate at which new cases, symptoms, changes, etc. develop for a specified time frame, such as hours or person years. I've seen the example of it being similar to miles per hour used as an analogy. So if I'm reading/understanding correctly (although I very might might not be), for the second example, the incidence rate would be .09 cases of recurrent VTE per person year (139 cases/1533 person years). The numerator is again the number of cases, but the denominator is a measure of time (such as the number of person-hours the participants were followed).

 

[aioo]

Quick answer, assuming I'm remembering/explaining my epidemiology correctly:

Cumulative incidence is the number of people developing a condition over a period of time. So in the first example, 32.7% of patients in group A and 34.7% of patients in group B developed the condition, symptom, outcome, etc. of interest over the specified time period (which looks to be 7 years). It's actually more of a proportion than a true rate. That is, it's the proportion of folks who developed the condition of interest over a specified period of time. The numerator is the number of cases and the denominator is the number of people followed. Cumulative incidence must always specify a period of time.

Incidence (or incidence rate) is the rate at which new cases, symptoms, changes, etc. develop for a specified time frame, such as hours or person years. I've seen the example of it being similar to miles per hour used as an analogy. So if I'm reading/understanding correctly (although I very might might not be), for the second example, the incidence rate would be .09 cases of recurrent VTE per person year (139 cases/1533 person years). The numerator is again the number of cases, but the denominator is a measure of time (such as the number of person-hours the participants were followed).

yes. but for the second example we have to answer if what they mean with ''rate'' is an incidence rate or cumulative incidence. So if the 28.6% and 1.9% is cumulative incidence or incidence rate. And the right answer is : cumulative incidence.

Because in the sentences before that they had mentioned person years and the fact that they say the adjusted 10yr recurrence.., adjusted 90 day ... I was also tricked to say that with the 28.6% and 1.9% they mean incidence rate, but the right answer is cumulative incidence...

I find this very confusing.

 

[WisNeuro]

Is this for a homework assignment?

 

[AcronymAllergy]

yes. but for the second example we have to answer if what they mean with ''rate'' is an incidence rate or cumulative incidence. So if the 28.6% and 1.9% is cumulative incidence or incidence rate. And the right answer is : cumulative incidence.

Because in the sentences before that they had mentioned person years and the fact that they say the adjusted 10yr recurrence.., adjusted 90 day ... I was also tricked to say that with the 28.6% and 1.9% they mean incidence rate, but the right answer is cumulative incidence...

I find this very confusing.

It's the cumulative incidence because it's still the proportion of people who developed the condition over a specified period of time (i.e., 10 years and 90 days), not the number of people who develop the condition per unit of time.

 

[aioo]

Is this for a homework assignment?

No its not. I use this as an example to understand the concept of the difference between cumulative incidence and incidence rate. Its really hard to talk about this without examples. And they come from an old exam that i have already made and know the answers..

 

[aioo]

So in incidence rate there must be a unit of time? But when I look at this example:

"In the week from 25 to 31 May (week 22), 12 on the 100,000 inhabitants were reported with influenza-like disease (IAZ) by the GPs participating in NIVEL Care Registry's first line."

the 12 on the 100,000 is here a incidence rate. But here there is no unit of time. And there is also a specific time period.


another example:

"In 2012, 477 men and 316 women died in the Netherlands through melanoma of the skin. This adds to 2.0 percent of total mortality from cancer in men and 1.6 percent in women. Melanoma of the skin is therefore a relatively rare cause of death, but increases strongly: from 470 deaths in 2000 to 793 in 2012. Relative to the population, mortality increased for this form of skin cancer from 3.4 per hundred thousand inhabitants in 2000 To 4.7 per hundred thousand
inhabitants
in 2012."

here are the 3.4 per hundred thousand and 4.7 hundred thousand incidence rates. But again there is no unit time used.

Thats why I struggle to come with something that I can use to differentiate these two types of incidence..

 

[smalltownpsych]

Unless you become an epidemiological researcher, you won't have to split these types of hairs. There were always a few concepts that I lagged on during my grad school education and it usually landed me an A anyway. Some of them I got later when I was teaching and some I just haven't had the need to know and still don't. The nice thing about this type of construct is half the time you'll be right on the exam.

 

[aioo]

Ok. I will just have to accept it and go on. Thanks everybody for trying to help.