Congratulations for the excellent initiative. I have worked already with health economics teams and never saw an explanation given so easy a straight to the point as most of your videos.
Glad you liked it! I will be doing more, just a bit swamped with work at the moment. Let me know if there are any particular topics you would want covered :)
@@TMSnowsill If you could do tutorials on how to implement markov models in R, it would be great. :) I'm a public health grad (health economics and policy). Will definitely be sharing your channel with friends. :)
Thank you so much for the series of videos! Just to confirm, the per-patient average cost and QALYs are for the 40 circles? Here circles can refer to any time period like year, month, as long as we set consistent unit for transition probability and payoff matrix? Thank you again for your excellent work!
Hi! Yes these are across all 40 cycles in the model. Cycles can represent any time period and you are absolutely correct that we need to make sure the transition probabilities and payoffs are consistent with the cycle length chosen 😊
Really a nice video! Your applications are clear and insightful. I just wonder how to conceptualize the initial amount of events (1000) in the first cycle. It is clear in the example of health economics. It is plausible that in the first observation, all statistical units are healthy. Less clear to me for the R-P-S example. Any further explanation is welcome
Thanks. The important thing for cost-effectiveness analysis is that the cohort size doesn't actually matter, because it scales both the numerator and the denominator of the incremental cost-effectiveness ratio. For budget impact analysis it is important and should be a relevant number for the policymaker, e.g., the number of new patients each year. In the R-P-S example it doesn't really mean anything - if we set the cohort size to 1 then we could interpret them as probabilities in each round.
These videos are extremely helpful! I am wondering if you could help answer a question I have on the inputs. How do you calculate your QALY based off of clinical trial data. Do you take the median life years of PFS and then multiply the utilities of PFS ? any advice would be helpful.
If I understand your question right that you're using a clinical trial data and then you want to apply QALYs to it. So you have to build your Markov model based on the health states in the clinical trial. For example, in oncology, usually clinical trials use (Progression-free survival, Overall Survival, etc.), so your model will be based on these health states. For QALYs, you have to look in the literature for articles that measure utilities for these states. Generally for diseases like cancer, diabetes, dyslipidemia, Alzheimer's, etc do have plenty of articles that measured health utilities and you can select from them (((((be careful, you should adjust the QALYs from the literature to the cycle length in your model,, for example if the article in the literature measured utility over 1 year and the cycle length in your model is 6 month, then you have to divide the utility score by 2))))). Then, based on the transition probability you run the simulation throughout your time horizon. Then, for each cycle, the number of patients in each state is multiplied by the utility score of that particular state. I hope this is helpful
Thank you so much for this wonderful tutorial. I'm still not sure how to come up with 57.3 and 21.1? Which numbers multiplied or divided by which number to get each of those values?
Hi! In the previous cycle there are 960 in the healthy state, 30 in the diseased state and 10 in the dead state. Of those in the healthy state, 96% will remain, and there is no way back to the healthy state, so we get 921.6 remaining in the healthy state. 3% of the 960 healthy (28.8) transition into the diseased state, but also 5% of the 30 diseased (1.5) transition to dead, so we have 30 + 28.8 - 1.5 = 57.3. For dead, we keep the 10 already there, plus add 1% of the 960 healthy (9.6) and those 1.5 from the diseased state, so 10 + 9.6 + 1.5 = 21.1. Hope that helps!
Excellent video; how about if you have a 3-state Markov model with a uniform starting distribution between the new intervention and comparator but with a varying parameter that affects one of the states across comparators.
Yes! Most of the time Markov models will have the same starting distribution for the different options (intervention and comparator), but certain parameters will affect how patients move through the health states and/or how those health states contribute to costs or QALYs. 😊
It depends on the time horizon for your economic evaluation. If it is lifetime, then you should keep running cycles until all patients are in the dead state, or essentially until adding more cycles makes no difference to the cost-effectiveness results.
Just made my thesis about this applied on Covid-19 with some real data and simulations. 3 states is quite simplified, but if we manage to increase further states and variables, also using probability functions for the matrix, we can actually achieve better results than government ones.
Thank you! At the moment I am not involved with any courses on modelling, although I do teach classes in health economics at the University of Exeter, UK.
Congratulations for the excellent initiative. I have worked already with health economics teams and never saw an explanation given so easy a straight to the point as most of your videos.
Thank you for your comments!
Thank you so much for this excellent tutorial. can't wait for the new ones
Glad you like them!
Woow, this is an amazing technic , simple and easy to understand.
Thank you! Cheers!
Excellent video! Couldn't thank you enough! I am grateful.
Glad it helped!
Thank you for this video. Can you do this for decision tree models?
Thank you so much! It is very helpful! But I'd like to ask if we need to consider discount rate in this process? Thanks a lot!
well done mate. amazing video
Thanks a lot!
Excellent video! Couldn't thank you enough!
Glad you liked it! I will be doing more, just a bit swamped with work at the moment. Let me know if there are any particular topics you would want covered :)
@@TMSnowsill If you could do tutorials on how to implement markov models in R, it would be great. :)
I'm a public health grad (health economics and policy). Will definitely be sharing your channel with friends. :)
Thank you so much for the series of videos! Just to confirm, the per-patient average cost and QALYs are for the 40 circles? Here circles can refer to any time period like year, month, as long as we set consistent unit for transition probability and payoff matrix? Thank you again for your excellent work!
Hi! Yes these are across all 40 cycles in the model. Cycles can represent any time period and you are absolutely correct that we need to make sure the transition probabilities and payoffs are consistent with the cycle length chosen 😊
@@TMSnowsill Thank you very much for your reply! Will continue following your work :)
Really a nice video! Your applications are clear and insightful. I just wonder how to conceptualize the initial amount of events (1000) in the first cycle. It is clear in the example of health economics. It is plausible that in the first observation, all statistical units are healthy. Less clear to me for the R-P-S example. Any further explanation is welcome
Thanks. The important thing for cost-effectiveness analysis is that the cohort size doesn't actually matter, because it scales both the numerator and the denominator of the incremental cost-effectiveness ratio. For budget impact analysis it is important and should be a relevant number for the policymaker, e.g., the number of new patients each year. In the R-P-S example it doesn't really mean anything - if we set the cohort size to 1 then we could interpret them as probabilities in each round.
These videos are extremely helpful! I am wondering if you could help answer a question I have on the inputs. How do you calculate your QALY based off of clinical trial data. Do you take the median life years of PFS and then multiply the utilities of PFS ? any advice would be helpful.
If I understand your question right that you're using a clinical trial data and then you want to apply QALYs to it. So you have to build your Markov model based on the health states in the clinical trial. For example, in oncology, usually clinical trials use (Progression-free survival, Overall Survival, etc.), so your model will be based on these health states. For QALYs, you have to look in the literature for articles that measure utilities for these states. Generally for diseases like cancer, diabetes, dyslipidemia, Alzheimer's, etc do have plenty of articles that measured health utilities and you can select from them (((((be careful, you should adjust the QALYs from the literature to the cycle length in your model,, for example if the article in the literature measured utility over 1 year and the cycle length in your model is 6 month, then you have to divide the utility score by 2))))). Then, based on the transition probability you run the simulation throughout your time horizon. Then, for each cycle, the number of patients in each state is multiplied by the utility score of that particular state. I hope this is helpful
Thank you so much for this wonderful tutorial. I'm still not sure how to come up with 57.3 and 21.1? Which numbers multiplied or divided by which number to get each of those values?
Hi! In the previous cycle there are 960 in the healthy state, 30 in the diseased state and 10 in the dead state. Of those in the healthy state, 96% will remain, and there is no way back to the healthy state, so we get 921.6 remaining in the healthy state. 3% of the 960 healthy (28.8) transition into the diseased state, but also 5% of the 30 diseased (1.5) transition to dead, so we have 30 + 28.8 - 1.5 = 57.3. For dead, we keep the 10 already there, plus add 1% of the 960 healthy (9.6) and those 1.5 from the diseased state, so 10 + 9.6 + 1.5 = 21.1. Hope that helps!
@@TMSnowsill It really makes sense. Thank you so much.
Excellent video; how about if you have a 3-state Markov model with a uniform starting distribution between the new intervention and comparator but with a varying parameter that affects one of the states across comparators.
Yes! Most of the time Markov models will have the same starting distribution for the different options (intervention and comparator), but certain parameters will affect how patients move through the health states and/or how those health states contribute to costs or QALYs. 😊
What changes when I can go from Diseased to Healthy state?
Thanks for the great vedio..
So nice of you
Can you please make a video on export and import of Agricultural products. Both the TPM and MCA
Hello, can you explain how to know how many cycles to run. Shouldn't we keep running cycles until all patents are in dead state.?
It depends on the time horizon for your economic evaluation. If it is lifetime, then you should keep running cycles until all patients are in the dead state, or essentially until adding more cycles makes no difference to the cost-effectiveness results.
Just made my thesis about this applied on Covid-19 with some real data and simulations.
3 states is quite simplified, but if we manage to increase further states and variables, also using probability functions for the matrix, we can actually achieve better results than government ones.
Glad it was helpful - great to see you have made something useful with it :)
Sir how I can calculate life years? Please help me
Great job! Do you teach a course?
Thank you! At the moment I am not involved with any courses on modelling, although I do teach classes in health economics at the University of Exeter, UK.