Spoonie Radio Ep 13: Joshua Grant of MENDUS
In Episode 13 I talk to Joshua Grant, founder of MENDUS--an online research lab dedicated to patient's exploration of their own health. Inspired by the quantified self-movement, Josh aims to garner ME/CFS research interest and help patient's self-experiment with this innovative project.
FULL TEXT TRANSCRIPT:
Dr. Craig: This is another episode of Spoonie Radio. I'm your host Dr. Courtney Craig.
My guest today is Joshua Grant who is the founder of MENDUS, which is the topic of today's interview. The motivation for this project came from his own struggles to self-manage symptoms of chronic fatigue syndrome. He decided to create a platform where users could come together, discuss ideas, and run their own research projects in order to find answers.
Josh is a neuroscientist that has worked in 8 different research labs in 4 different countries on various topics ranging from visual perception and cognitive pain modulation to human affiliation. He is currently a postdoctoral fellow at the Max Planck Institute in Leipzig, Germany.
So welcome to the show, Josh!
Josh: Thanks, nice to be here!
Dr. Courtney: Yeah, thanks for being here. So just to start off, I would like to hear a little bit about your background and how your background evolved into the creation of MENDUS--and this idea for MENDUS.
Josh: Well it's really like you just read. I mean I think I've had chronic fatigue syndrome for a long time. I had no way to know that for sure but that's my guess. I have been tired since my teens.
I think being single most of my life I was able to sleep whenever I needed to and probably manage myself that way. But the last couple of years having a family now I am not able to do that and I started to break down quite quickly. So I went to doctors and this was the conclusion – chronic fatigue syndrome.
And being a researcher--I am sure that others that aren't researchers do the same--I just kind of poured over the Internet looking for possibilities of things to try because basically they told me at the hospital to take ginkgo biloba and sleep or something like that.
I looked at all the options and there was so much. And through the years even without a diagnoses I had tried stuff to pick up my energy levels or in order to stop pain or whatever, and for some reason I never did anything systematically. I would take a supplement for a week and some days I would miss it, I wouldn't pay attention to how much I should have taken her how much I did take or I didn't record my symptoms. So you get a vague estimation of whether it works or not but there could be so many other things that are involved like how much exercise you did. That it's really a crappy way to evaluate whether something works and being a scientist I should've known better a long time ago.
So I started this site more or less just to keep track of my own things. I made these little, which are nothing more than a Google form, little boxes that I could click and put a rating of my fatigue or pain for a particular day.
And then I decided I would just kind of choose something, whey protein or whatever and read the literature on it to base the dosage that I would take. And then daily or multiple times a day, make ratings of my core symptoms and then actually try to run some statistics on it overtime and see if something statistically or scientifically actually meaningfully had a change.
And then that's still a valid idea and I did that. But ultimately I mean it's not of much use to anybody but me, and even then it's hard to say in certain situations whether data from a single person is real. I mean I guess it's real but it can't be extrapolated to everybody if it's only from a single person.
So I decided to open it up to whoever might be interested. I didn't really have any idea how it would shape up what it would look like. I had no idea of the time investment. It's taken a hell of a lot of hours.
Dr. Craig: Yeah, absolutely. So would you consider yourself a biohacker or part of the quantified-self movement? Would you identify with that?
Josh: To be honest, I came up with the idea completely independently. I told some friends at the Max Planck Institute about it and asked if they were aware of anything like it and the one guy is like – oh, there is an entire movement called Quantified Self and I said, “Whoa, really?" So I looked into it and yeah, I mean it's practically the same thing except they don't really attempt to merge these individual data sets into groups and statistics. So I guess I am kind of spanning both of them.
Dr. Craig: Yeah, it's a unique way to look at it. So basically tell us how does MENDUS work? You go to the site and what do you do?
Josh: Some of how it works is well-established now. Other ways it's kind of evolving because I didn't really have a clear destination for it. So the idea is people would come, it's broken up into communities based on the disorder – so chronic fatigue or ME, I don't know how to pronounce the other complicated word. Fibromyalgia. We have a multiple sclerosis community with one person in it, that was a person that actually asked me to create it and I have shunned by any attempts to advertise this for those patients.
Then a B12 deficiency group – pernicious anemia. We are open to making new communities for people that are interested. It could take a while now because I am pretty swamped. But people would come to the site; there is a separate registration procedure for each disorder, they have their own form more or less.
The reason I did that is I need to keep things completely anonymous and still be able to track everybody's data if they submit things at different times. I need to issue them an ID number. And to do that, without any computer knowledge, I came up with a rather strange procedure with copying numbers off of a little clicker on your page which guarantees that I will never have a double of a member.
So then they would come to the site. There is a little button called "Communities" and they could go to their particular one or ask me to build a new one. From there, all of the pages are duplicates of each other. So there is chronic fatigue homepage has an image it says chronic fatigue and then on the right side there is a few buttons, and these are the main links from the page and then there is a form on that main page.
So on the right side these buttons include "Register" that would take them to little form that leads them to the steps to get their ID number.
As I said before, there is no names involved, I never ever ask for names. The most I need is an email address. And then the other buttons on that line once you registered I think there is preliminary survey button. So for each different disorder I have tracked down a clinical instrument that people can fill out. The chronic fatigue ones are quite long. It's the DePaul Symptom Questionnaire, the fibromyalgia one is pretty short.
And the idea here is that we can get a general estimation about where a person is on the disorder spectrum in terms of the severity of their symptom and things like this. I think that information might become important later when we have more data on different types of studies to revert back to where they were when they started and maybe there is a subgroup where one thing works and another subgroup where another thing works.
But it also allows the person to see their own data in relation to everyone else on the website and in relation to the original publication where the instrument came from. So for the chronic fatigue one, Leonard Jason, this is his questionnaire and he validated it in the Solve ME/CFS cohort of people. So I took their values from their table on their paper and made a graph of it, a bar graph and put our people against it so that they can compare where they fit with this Solve CFS group. It turns out to be exact almost, it's identical, it's unbelievable how close those symptom patterns are.
Josh: So they can do the preliminary survey. We ask people to do that right away or before they do any studies. It's not totally mandatory of course but nothing is at MENDUS, but we encourage that.
And the next button down is "Studies" and here you will be taken to a page where there is a table with…I’ve just given studies of very general name like the coenzyme Q study and the whey protein study. And it lists all those studies.
There is current status whether it's an active study or it's still waiting to get some more interest, and then a number of people that are interested in it. So currently most of the activities has been through the chronic fatigue group. So on their table there is I think four, five active studies; a diet study, the D-Ribose study, I just started the Q 10 study.
Dr. Craig And how many people will need to be interested before you say, “Hey, let's do this?"
Josh: Originally I had thought maybe 20 but I waved that because there was 20 and I was ready to go. So right now there is no strict rule and it's kind of just gone by my own thinking.
There are 2 studies on the chronic fatigue board that have been there for a while and are probably important studies but they are really complicated and I have kind of put them off because they are complicated and there wasn't enough traffic to the website yet to really warrant that kind of effort I think. So instead, those ones I am referring to are the vitamin B 12, or any B vitamin study for chronic fatigue.
The reason that one is complicated is because it might involve injections of the B12. That's really the best way to do it. But I mean in Germany here is totally fine. You can buy everything over-the-counter but I don't think in the United States or Canada or the UK you can do that.
Dr. Craig: You definitely can't do that here.
Josh: Yeah, it would be much tougher. It will involve people going to their doctors and I want to cover myself and make sure everything is accurate so I need a nice consent form with information for people to read. So that one has been held back a little bit. Then there’s an anaerobic exercise study based on I think the Nancy Klimas’ protocol that was the idea.
But again if we are asking patients to try to do exercise which could hurt them, I decided it's probably best to do this when it's really, really well thought out and I haven't had the time to do that. Whereas this Q 10 study and the whey protein and the chocolate study, these are very simple to put together, it's just a matter of telling people how much they should take and these are things that are supposed to be completely safe for anyone to take so there is not no risk that I see is involved.
So a question of how many people need to be involved, there were really no people involved in the Q10 study. Well I can't really say it because it's not official, but there will be a study coming that we needed a control condition for and that's what the coenzyme Q study in itself will serve as a control condition for another one that will be coming shortly that could be quite exciting as well.
Dr. Craig: Yeah, I know one of the ongoing studies I think that has maybe the most participants is the CFS diet study where you are having patients track each meal and using an app to record their food intakes. So how do you go about designing that with clean data with all these possible confounders?
Josh: Well, that's super difficult. I guess technically impossible. There is a lot of confounds, people have been pointing them out. And like I said I kind of went into this not really sure what was going to happen and just a lot of things that have happened that I don't think I could've anticipated.
In the diet study, we are tracking 30 nutrients. All of the tracking is done externally by another website and people just send me their outputs, so that's okay. I have had to make sure that everyone put things in the proper order so that when I merge these things into a huge spreadsheet we have all of the proper things in the proper columns.
Shortly after I started I realize like you are saying, there is so many things that could influence the data so I had people start tracking if they do any exercise. The idea there was just to get a single kind of 1-10 measure each day of how much activity they have done, and then statistically we can put that into the equation and it will in a sense tell us what the effect is when that exercise is at constant level. We did that also with sleep. So their sleep: number of hours and the quality of their sleep.
Now the problem with that, that's great and that would definitely help the results an immense amount. Let's say you track your diet, you ate the exact same thing for seven days but on day four you ran a marathon. I mean clearly something is going to happen to your pain and your fatigue ratings because of the running not because of your diet.
So tracking that should be able to help us remove that from the equation and still get a good estimation. The problem then is to try to offer the participants outputs, like graphs, of the results accounting for those, all of those things is incredibly difficult. I have already surpassed my own stats knowledge and had to learn new procedures in order to analyze this stuff.
I have 10,000 columns of diet data. My Google spreadsheets are crashing my computer. I have to think of a new way to do it now.
Dr. Craig: Have you noticed any patterns or insights? Are you able to talk about that yet?
Josh: For me, I didn’t find anything. There is probably a good reason for that which we could get into later if you want, but for a couple of the ladies room studies it actually turned out a lot. There is one particular woman who has been kind of addicted to it, I am trying to convince her to maybe go off of just the basic tracking and maybe test something. She has been doing it, I think she has done it for 80 days now or something.
So we looked at her data yesterday. It appears that her protein levels are too low. I mean with 80 data points she has plenty of data to find a good balance. And so with the energy levels and more if she is more protein she has more energy, more mental clarity and less pain and all of those are statistically significant as well as when I control for her sleep and her exercise. And she has lost apparently 40 pounds. I don't know what she has been doing to lose weight but she has also lost 40 pounds in the process. So for her it's fantastic and that makes me happy and kind of gives me a drive to keep pushing forward on it.
I think some people maybe diet isn't a factor or there is something confounding it that's in there that we are not controlling for like myself. I don't know why. I expected to find great things and for me it was a bit discouraging. There’s patterns in there that aren't strong and I wasn't able to really figure out what they meant but perhaps with the help of a dietitian – I mean I don't really know much about diet.
Dr. Craig: So that brings up an interesting point because when I first heard about MENDUS I am like, “Oh, this is great! We will get the studies created and then it will spark interest in bigger research institutions, academic institutions to create more refined blinded controlled studies."
But what you are describing here is one person benefiting from the study and there is a lot of value in that too. So just being in this community can actually help individuals as well as the entire community by stemming interest for maybe research projects that can be funded and published, right?
Josh: Yeah, I mean the diet study is different than all the rest so far. It's not really even a study in the proper respect because we are not manipulating anything. It's just, I guess it is called an observational study. We turned it into that just personally with me and two of the ladies doing it through email. I haven't figured out a good way to implement this in the website. So what they’ve done is they stopped tracking everything after a while.
They look at their patterns and decided, “Okay, I'm now going to try reducing this." In order to evaluate whether that does anything, they have to track a different, I just add a "B" behind their ID, and I have two sets of data more or less and you can compare them.
And that worked quite well. We found out the one lady was more or less taking vitamins when she was in pain and it led to a weird effect but it looked like the vitamin caused pain. We don't really know, I doubt that that's the case.
So yes, for individuals, the diet study in particular could be really, really valuable. The other ones…I mean yes, I am providing people's results person by person. They can enter their ID and they will see their pain from week one to week four and am sure that that's valuable to people. As I said in the beginning my own approach was just kind of haphazardly take something and then, “Yeah, I think I feel better." Well this will be a way for them to see; yes, I felt better over the four weeks. The bar graph shows that. Or my pain went down. So I think it's valuable for the people.
You can't really do stats on an individual level with all the other studies. You can with the diet study because they have hundreds of data points. Each day they are eating 30-40 items and you have that for four weeks and making symptom ratings 8 times a day sometimes so they have a huge data set and it's okay to do stats with that but you can't with four points for a single person.
So the D-Ribose study and all of those, it will show you your pattern but you can't really say that's statistically meaningful pattern. I don't know if people care to be honest. Scientists care and doctors I am sure, but I don't think the people care so much. If they see that it made them less tired I think that would be good for them. But the cool thing is that you could still do the regular study, the regular stats, the regular group things. I mean it's not quite as clean as the study, you kind of mentioned that before but the D-Ribose, this was the first one after the diet study.
I felt like because the primary goal was to help people, individuals, to see whether it works for them that there was not so much point in restricting them to what product they use. So they could use any kind D-Ribose, they could get their hands on. That then enters a huge amount of variability. I mean perhaps someone got mixed with magnesium, maybe, maybe it was a magnesium that did something.
Josh: But since then I have tried to refine it and get people trying more specific things. So for the coenzyme Q10 I am asking people to do a very specific product. If they don't want to, that's okay too I am going to turn anybody away. But I ask them to record it so that at least I know what they took.
Dr. Craig: So as a scientist and having a scientific background, how do you think the scientific community at large will view this project? Or have you already gotten feedback maybe from other ME/CFS researchers out there? Are you getting respect here or is there a lot of skepticism? What do you think the general consensus is with your project?
Josh: I got a bit of both to be honest. There is people that will just claim that this isn't science. Well okay, that's fair enough. I mean I am the first to admit it's not. There is no way we can control everything that we can in the lab. I had a detailed discussion with a friend of mine who is a professor in the UK, and he brought up a ton of good points, not all of which I had thought of was some of them.
And my answer was well, you have to think about what the point of it is. I am not necessarily trying to publish papers with these studies. Some of them may lead to that; if a journal will take it I don't know yet but.
Everyone is doing it themselves there is just automatically going to be a ton more variability in the data. So it's a question of how I want to restrict it in order to keep it tighter, more controlled study but again and it is going to help less people because less people will get involved. So I've kind of erred on the side of helping people and the statistics will suffer a bit. And if anything it might just lead other researchers in more controlled environments to actually do a real science study.
It is of note, I'm trying to get more and more controlled all the time. For the coenzyme Q10 study, people can actually do a placebo condition, and help us get a healthy control group. And all the basically need to do is stick with the products I'm suggesting. It's nothing special. I didn't pick it because it was supposed to be better than anything else. I picked it because the gel caps are unmarked.
So it's Now Foods, they are available all over the world, they are fairly cheap but yet a high quality Q10. They have both ubiquinone and ubiquinol in the proper dosage, and they have vitamin E caps that are identical. So they could quite easily by two bottles of these things and have someone switch them up for them and put them in an unmarked container and then do their own placebo.
I have someone doing this in the D-Ribose study. She found a sugar substitute that she had never tried before, she had never tried D-Ribose and so she did her own placebo study.
And so the other one is to ask someone, a friend that is relatively healthy, to run through it with them. I mean the D-Ribose I don't think will hurt a healthy person, I hope not. And Q10 the same thing, it is probably good for everybody. So if you were to just get your partner if they are relatively healthy or a friend and they run through the study with you then they would be able to serve as the healthy control group.
I mean I think we can get more control it's just a matter of creativity and being open to the fact that it's not going to be perfect.
Dr. Craig: That's very creative for sure. What sort of instruments are you using? I mean there is so much great technology now, there is apps being developed to self-track, being born out of the quantified-self movement. Are patients using any of these technologies?
Josh: In the diet study they are. I mean they are using the Spark People website. I think they have apps for the phones, for the mobiles as well.
For the other studies I haven't done that yet. I don't know a lot about that to be honest. I have had cell phones the last maybe 15 years but only ones without big screens, it was just out of necessity. I do have one with a big screen now but again I haven't really used it for apps. I know there is a ton out there and I have looked into it for the diet study but not much for the others.
I was thinking of trying to build one to be honest for the website because right now I am just using kind of something very crude, it's like a Google form.
The nice thing about that is that it goes to a spreadsheet where I can write in formulas and things like that and the analysis will more or less do itself and so my time investment is greatly diminished.
If this thing actually works meaning MENDUS, I mean I have no idea I am just a guy running this stuff, it's not like it's a huge company or anything. I made €18 in five months through Google ads or something like that, €18. So if people are thinking the ads are generating a lot of money, that's not the case.
Anything that it were to make will be put straight back into it--into hiring someone that could help with apps and properly making these forms to do the studies, maybe get a statistician to help with the statistics. Yeah, there is so much work to it that I didn't envision that I am doing almost everything myself.
Dr. Craig: So that sounds like part of your bigger goals. I mean what is the long-term picture of MENDUS look like? Do you have any other studies that you want to see get started that haven't been put up on the site yet?
Josh: Well the anaerobic exercise one…and this ties in with your last question. This would involve tracking probably heart rate and maybe heart rate variability. Perhaps even oxygenation of the blood if you can do it. And I know it's possible but I don't know if little devices for a single person are available. I know you can do it at the lab quite easy.
So this was an idea is to get the tracking, there is tons of exercise trackers for heart rate and heart rate variability and stuff like that. We are going to work that into the anaerobic exercise study. I really would like to see that one go. I think that's important. I think especially for chronic fatigue syndrome where it's such a crappy position. You need exercise to kind of be healthy but you can't do exercise.
It's future? No idea really. I am willing to keep going with it, it's taken a lot of my time. If more people are involved fairly soon and if people were to buy their supplements for the studies through the links that I provide then it's set up so that a portion of the sales go to the website. That would help immensely because then someone could be hired to help and then there is no doubt that it will continue. What I would try first is a programmer or a webmaster programmer that they could clean it up and they could run more efficiently if there is more people trying to deal with my registration system. It's terrible.
In the future, who knows? I mean so far the feedback has been extremely positive. Doctors, a few researchers, there has been more positive than negative even from researchers.
So some members have suggested genetic studies if people were to share their 23andMe or equivalent data than that could go in. I mean that requires a lot of knowledge of genes and interactions and that's not my strong suit. I have done a bit. So I think it would require, if it continued and got bigger and bigger, it's going to require not only the input of other people in terms of programming but probably other scientists. I wrote that into somewhere in the mission statement.
Ideally if we are doing genetics where you could have someone that's an expert in genetics involved. Now whether I could get someone I don't really know. I mean I would think that a person in the lab that studies genetics and chronic fatigue might be interested in this as a platform kind of for preliminary data even if they don't consider it possible to publish it. If they worked with me or us they could in a sense test out their idea before they go full guns on it. It might help them get grants if they have something some preliminary data to show a granting agency.
Dr. Craig: Yeah, exactly and MENDUS provides a great jumping off point for a lot of potential researchers and we have other great resources available too like the BioBank. So if you pulled a bunch of preliminary data, particularly genetics, you could then apply for a grant through a bio bank and use all of those samples to compare. I think it's a great opportunity for researchers out there.
Josh: Yeah, that was one of the main ideas, to kind of just stimulate ideas and have them coming from the patients themselves. I mean they are the people that are feeling these things. When I read about the study on Cort Johnson's site I can't remember the name of the researcher, she is in the UK looking at the acid in the muscles of chronic fatigue syndrome patients. About a year ago they found that there is 20 times the amount of acidity in the muscles of chronic fatigue patients for the same amount of exercise as the healthy control subjects. This is totally insane!
Dr. Craig: So hopefully the listeners today can help with the effort at MENDUS. And Josh, can you direct our listeners were to go to sign up to get involved with some of these studies or to introduce some of their own ideas?
Josh: If you Google "MENDUS" we should be on the first page. It's M-E-N-U-D-S. The website itself is www.MENSUS.org . We also have a Facebook page that has all of the things any new study will be put out there and links to the main site.
On the main page there is links to the different communities and the different studies. Some people have said that the website is a bit hard to navigate, others have said it's great so it must just be individual differences.
Dr. Craig: I found it pretty easy to navigate. I signed up a couple months back.
Josh: Cool. Everybody is welcome, new communities, it might take a while but they are welcome.
Dr. Craig: Well, I hope this episode certainly gets a lot of new patients on the site and I'm looking forward to seeing more good ideas of studies coming involved and see what happens. This is new but I think it's a great project and thank you so much for using your talents as a researcher to help this community. That's awesome.
Josh: Well, it's my pleasure and thank you for bringing attention to us.
Dr. Craig: Okay, with pleasure. All right! That's it for another episode of Spoonie Radio. Be sure to subscribe over on iTunes so you never miss an episode. Until we meet again Spoonie Radio signing off.