Yes but without proper cohort selection the whole survey is meaningless. You can't mix:
1. I feel a bit tired - oh I must have "Chronic Fatigue". I get enormous help from X,Y,Z lifestyle factors (e.g. switching jobs).
with
2. I have a debilitating neurodegenerative disease. I have been unable to work for a decade. There aren't any "lifestyle" factors that I haven't already tried. Some help, but none affect the underlying disease. I need serious drug/nutraceutical treatments to make a dent in my condition.
The problem is that group 1 have so far monopolized "CFS". Group 2 are too sick to advocate for themselves and have been left to rot.
We need to separate the two groups, and give them different names. It's not fair that group2 should have treatments denied based on surveys conducted by group1. E.g. I know of someone who was denied a drug in the UK because of such a treatment survey conducted by the ME Association.
I think lifestyle factors need to be separated from biomedical treatments, especially given the history of this disease and everything that has gone on for the last 25 years.
Hi, Garcia.
I understand your argument, and I did pass it on to Daniel Reda. Here's what he wrote in response:
"Regarding the point about not referring to lifestyle changes as a type of treatment, we are not trying to shape the perception of any condition in any way. Our intent is merely to gather and list, as comprehensively as we can, and within appropriate categories, the types of interventions and actions people take in their attempts to feel better. Virtually all health conditions include such lifestyle and dietary changes as interventions. It's unfortunate if some people assume that if rest is the highest rated treatment for CFS, then it must not be a "real" condition, but we cannot control that. More importantly, we would not want to deny CFS sufferers the information about rest being so helpful."
So long as the people who post for CFS on "crowd-sourcing" sites are self-selected, which is inherent in this type of a survey, there will always be ambiguity in whether or not they actually have the more serious disorder you have described for "group 2." However, these surveys will probably become more prevalent, because crowdsourcing seems to be gettting more popular.
Perhaps the best way to work toward correcting this problem is to encourage more of the people in what you have called "group 2" to post their experiences on CureTogether. If more people from this cohort post data indicating that they did not find the lifestyle changes to be effective in their cases, it will lower the calculated "average effectiveness" of these actions, and they will drop down in the comparative ratings of average effectiveness. Of course, this assumes that there will be treatments that will be found helpful enough to rise above the lifestyle changes. I'm optimistic that methylation treatments are one type of treatment that have that potential, but time will tell.
I also conveyed the issue of whether "popularity" is appropriately named on the CureTogether site. Here is Daniel's response:
"Regarding the chart, by popularity we refer to the % of respondents who have tried a treatment, regardless of whether they continue to use it today. The point of the chart is to show that often, less effective treatments are tried by more people than more effective ones - which is a measure of needless suffering. If there was good information out there about what worked and what didn't, we'd expect the dots to be tightly clustered around a smoothly increasing trend line. With more poorly understood conditions, the dots tend to be more scattered. Perhaps we can do a better job of explaining this on the site. I'll think further on this, and welcome any suggestions you may have on how to better label the x-axis."
Perhaps the X axis could be labeled "% who have tried the treatment."
I think the ultimate solution to the group 1--group 2 problem is to develop better diagnostic criteria, with good biomarkers. The current case definitions and diagnostic criteria are too muddy to separate these two groups.
Best regards,
Rich
