Re: bearing: I'm glad you did include something like this. Another angle would have been protective non-action factors, but your wording seems more solution-oriented.
Re: EQ-5D-5L (or something similar): I agree that the suffering and health-related quality of life or health utilities only partially overlap and that some dimensions will have more predictive power than others. What I would be interested to know is how big the discrepancy is between reported suffering and a generic instrument. Are people who suffer greatly reporting middling problems on other commonly used instruments? or How much does conventional QALY tools under-report extreme suffering? It could be another argument for why and how existing tools and metrics are inadequate.
One thing I struggle with in this area is how to think about temporal aggregation. Is one hour of suffering on two days worse than two hours of suffering one day? I am nonetheless glad to see OPIS trying to map out a 'global burden of suffering'.
Intriguing. I am not too surprised that psychological conditions can cause extreme suffering but I think the mapping of the suffering landscape here is great!
Some quick initial thoughts on how to proceed:
Re. your discussion about the top end of a scale in the third last paragraph: I generally do not like using numerical scales for things that do not have a meaningful numerical interpretation. I prefer the parts where you treat the suffering score as a categorical variable fuelled by some semi-numerical heuristic. I also like the suggestion of defining 10/10 as "literally unbearable" There is something qualitatively (and maybe morally) different about literally unbearable states, i.e. worse than death. It is also much harder to distinguish between degrees of unbearable suffering than it is to distinguish between degrees of bearable suffering, and I think it can be a good enough first approximation for now.
Other suggestions: I would also like to see the inclusion of EQ-5D-5L or some other generic health instrument and a correlation plot between the health utilities of the self-reported health states and the suffering score (even if it goes against my previously stated distaste for treating the suffering score as a numerical variable). Another thing that is a bit besides the scope of this study: A short-form question about how they bear this suffering.
Re the comparative approach in the second-last paragraph, It would also be interesting to see. It would, however, not estimate anything theoretically analogous to DALY weights unless you add some trade off-based exercise.
Given its gravity, I find this question underexplored and current tools (QALYs, WELLBYs, YLSSs etc.) either underdeveloped or inadequate. At the same time I think answers should fit into existing frameworks when possible, since these tools often form a cornerstone, or are heavily implied, when doing cause prioritisation.
There are at least two things I would like to know more about:
When does the transition from positive to negative lives happen? What characterizes this shift? Some of the above tools have a defined neutral point, but in practice, evaluations around it are somewhat distorted.
How far in the negative is it possible to be compared to the positive?
I think it should be possible to at least get some empirical anchors tied to commonly used frameworks, preferably more than one.
(And I have some ideas for how to do so, but it doesn't seem like there are many funding opportunities, so I'd probably want $500 000 for myself to pursue them, or maybe just $50 000 to do some small tests)
I get the impression posts along these lines pop up from time to time, see e.g. the one ben linked below. Personally, Id like to see way more (cross) cause prioritisation than today, but this is constrained by funding rather than interest or talent.
Not exactly my field, but I had a quick chat with a conveniently located senior researcher at work. This is a synthesis of our thoughts:
Summary:
Might be a bit ambitious, but doable if you are well prepared.
Feasibility:
The number and availability of of informants will be crucial. Do you already have contacts or a way in? I frequently see people (including myself) grossly underestimate the time it takes to collect data. How many informants? If you plan to publish the results in a peer-reviewed journal, aiming for more than the minimum (I don't know, 8?) is probably wise.
Transferability:
I would imagine some findings would be generalisable to similar countries at least.
Methods:
What are your options? Would a survey be possible? Need more people but is maybe easier to analyse. Thematic Analysis is a good pragmatic approach. Because of this pragmatism however, your current description lack some detail on how you will actually do the analysis (except the general steps). For example, how will you relate the data to your concept model? You won't have much time to figure this out during your 10 weeks. Especially if you have never analysed interviews before.
How to conduct interviews:
If you know someone who already is doing a similar analysis, see if you can sit in on an interview, working session or supervision session and observe. It helped me the time I wrote a thematic analysis paper (but I will not offer or trust my own skills here)
Pitfalls:
Just want to reiterate that data collection sometimes takes much more time than anticipated, and you should expect to spend 80% of your time on analysis.
It seems like an important crux in your analysis is quantifying the intensity of CH.
I'd like to point out that QALYs as a metric is not mentioned here. In the QALY-paradigm, the utility weights are anchored at 1 = full health and 0 = dead. Importantly, negative utility weights are also theoretically possible. For example, an utility weight of -10.0 would imply that removing one person-year of CH would be equivalent to 11 QALYs (which is equivalent to the absolute prognosis loss for one person with chronic migraine, according to one random report from the Norwegian Medical Products Agency i just dug up). However, current methods for eliciting negative values are imprecise and somewhat arbitrary. I've been thinking about whether developing better metrics within the QALY paradigm can be useful, since it is more widely adopted. CHs would be the perfect example case. Curious to hear if you have any thoughts on this.
Good points! Some follow-ups:
Re: bearing: I'm glad you did include something like this. Another angle would have been protective non-action factors, but your wording seems more solution-oriented.
Re: EQ-5D-5L (or something similar): I agree that the suffering and health-related quality of life or health utilities only partially overlap and that some dimensions will have more predictive power than others. What I would be interested to know is how big the discrepancy is between reported suffering and a generic instrument. Are people who suffer greatly reporting middling problems on other commonly used instruments? or How much does conventional QALY tools under-report extreme suffering? It could be another argument for why and how existing tools and metrics are inadequate.
One thing I struggle with in this area is how to think about temporal aggregation. Is one hour of suffering on two days worse than two hours of suffering one day? I am nonetheless glad to see OPIS trying to map out a 'global burden of suffering'.
Intriguing. I am not too surprised that psychological conditions can cause extreme suffering but I think the mapping of the suffering landscape here is great!
Some quick initial thoughts on how to proceed:
Re. your discussion about the top end of a scale in the third last paragraph: I generally do not like using numerical scales for things that do not have a meaningful numerical interpretation. I prefer the parts where you treat the suffering score as a categorical variable fuelled by some semi-numerical heuristic. I also like the suggestion of defining 10/10 as "literally unbearable" There is something qualitatively (and maybe morally) different about literally unbearable states, i.e. worse than death. It is also much harder to distinguish between degrees of unbearable suffering than it is to distinguish between degrees of bearable suffering, and I think it can be a good enough first approximation for now.
Other suggestions: I would also like to see the inclusion of EQ-5D-5L or some other generic health instrument and a correlation plot between the health utilities of the self-reported health states and the suffering score (even if it goes against my previously stated distaste for treating the suffering score as a numerical variable). Another thing that is a bit besides the scope of this study: A short-form question about how they bear this suffering.
Re the comparative approach in the second-last paragraph, It would also be interesting to see. It would, however, not estimate anything theoretically analogous to DALY weights unless you add some trade off-based exercise.
"What does 'negative lives' really look like?"
Given its gravity, I find this question underexplored and current tools (QALYs, WELLBYs, YLSSs etc.) either underdeveloped or inadequate. At the same time I think answers should fit into existing frameworks when possible, since these tools often form a cornerstone, or are heavily implied, when doing cause prioritisation.
There are at least two things I would like to know more about:
When does the transition from positive to negative lives happen? What characterizes this shift? Some of the above tools have a defined neutral point, but in practice, evaluations around it are somewhat distorted.
How far in the negative is it possible to be compared to the positive?
I think it should be possible to at least get some empirical anchors tied to commonly used frameworks, preferably more than one.
(And I have some ideas for how to do so, but it doesn't seem like there are many funding opportunities, so I'd probably want $500 000 for myself to pursue them, or maybe just $50 000 to do some small tests)
I get the impression posts along these lines pop up from time to time, see e.g. the one ben linked below. Personally, Id like to see way more (cross) cause prioritisation than today, but this is constrained by funding rather than interest or talent.
There are many judgement calls in this type of work. A new moral weights project will give useful information about how robust results are.
Not exactly my field, but I had a quick chat with a conveniently located senior researcher at work. This is a synthesis of our thoughts:
Summary: Might be a bit ambitious, but doable if you are well prepared.
Feasibility: The number and availability of of informants will be crucial. Do you already have contacts or a way in? I frequently see people (including myself) grossly underestimate the time it takes to collect data. How many informants? If you plan to publish the results in a peer-reviewed journal, aiming for more than the minimum (I don't know, 8?) is probably wise.
Transferability: I would imagine some findings would be generalisable to similar countries at least.
Methods: What are your options? Would a survey be possible? Need more people but is maybe easier to analyse. Thematic Analysis is a good pragmatic approach. Because of this pragmatism however, your current description lack some detail on how you will actually do the analysis (except the general steps). For example, how will you relate the data to your concept model? You won't have much time to figure this out during your 10 weeks. Especially if you have never analysed interviews before.
How to conduct interviews: If you know someone who already is doing a similar analysis, see if you can sit in on an interview, working session or supervision session and observe. It helped me the time I wrote a thematic analysis paper (but I will not offer or trust my own skills here)
Pitfalls: Just want to reiterate that data collection sometimes takes much more time than anticipated, and you should expect to spend 80% of your time on analysis.
Thank you for a very interesting read.
It seems like an important crux in your analysis is quantifying the intensity of CH.
I'd like to point out that QALYs as a metric is not mentioned here. In the QALY-paradigm, the utility weights are anchored at 1 = full health and 0 = dead. Importantly, negative utility weights are also theoretically possible. For example, an utility weight of -10.0 would imply that removing one person-year of CH would be equivalent to 11 QALYs (which is equivalent to the absolute prognosis loss for one person with chronic migraine, according to one random report from the Norwegian Medical Products Agency i just dug up). However, current methods for eliciting negative values are imprecise and somewhat arbitrary. I've been thinking about whether developing better metrics within the QALY paradigm can be useful, since it is more widely adopted. CHs would be the perfect example case. Curious to hear if you have any thoughts on this.