search

zward / Amua

An open source modeling framework
GNU General Public License v3.0
33 stars 11 forks source link

A unit time for each cycle is 1.5 month but a result using year (i.e., 12 month) as the denominator #50

Closed yangszuchun closed 4 years ago

yangszuchun commented 4 years ago

Excuse me! If the unit time for each cycle is 1.5 month (6 weeks) but I would like to get a result using life year or QALY (i.e., 12 month instead of 1.5 month) as the denominator, how to adjust the property Amua? Thank you very much in advance!

zward commented 4 years ago

You should be able to use any cycle length, as long as you convert the transition probabilities and rewards. Annual transition probabilities need to be re-scaled to match up with the new cycle length - you can use the probRescale function. To calculate QALYs you would change the rewards from 1.0 to 1/cycles per year. So moving from a one-year cycle to a one-month cycle you would divide all of the rewards by 12, for example.
Does this answer your question?

yangszuchun commented 4 years ago

Thank you for the kindest reply! I am sorry that I did not explain very well. I used 1.5 months as a unit, and the ICER was $37140/1.5 months (as below and attached). CEA Results (Ratio of Means):Strategy            Cost        QALY    ICER      Notes   -------------------- ----------- ------- ---------- --------Chemotherapy        192743.8503 11.7906        --- BaselineNivolumab+Ipilimumab 494098.828 19.9045 37140.3972 But I would like to have a result of  37140x8/12 months (like that in the scatter plot). How to solve the problem?Again, thank you for your kindness! Best,Szu-Chun 在 2020年3月16日 星期一 上午08:25:04 [EDT]， Zach Wardnotifications@github.com 寫道：

You should be able to use any cycle length, as long as you convert the transition probabilities and rewards. Annual transition probabilities need to be re-scaled to match up with the new cycle length - you can use the probRescale function. To calculate QALYs you would change the rewards from 1.0 to 1/cycles per year. So moving from a one-year cycle to a one-month cycle you would divide all of the rewards by 12, for example. Does this answer your question?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

zward commented 4 years ago

Hi Szu-Chun, I'm still not sure I understand what you are trying to do. I don't see a scatter plot either. Do you want to send me a copy of the model (zward@hsph.harvard.edu) and describe what the problem is?