Something magical is going on here

[momto2pandas]

p.s. I should mention that the "interested parties" with which the data would be shared would be restricted to medical/scientific professionals, and the analysis would have to go through an analysis specialist, else the potential for poor analytical techniques leading to wrong conclusions or for the data to be used in the wrong kinds of ways.

 

 

Much of what you describe in the beginning is why I left academia! It's a painfully slow process, and you didn't even mention the parts about politics or the fact that what's vogue in any particular era is what gets funded....or the fact that research findings can be suspect due to publication biases, etc. Pharma moves more quickly and easily (and with more regulation), but of course mostly purues profit-oriented questions.

 

I would like to expand further on the "new" model. I think it would be great to take a formal survey approach - allowed any patients with verfied diagnoses to "opt in" to email mailing lists, and then each time a new question arises (by the scientific managers) or new data is created (by the patients), a survey expert formulats appropriate questions and patients can follow a link to enter their information into a well-designed survey/data collection instrument. Someone expert in data analysis then mines the data at appropriate intervals to look for patterns and generate/test hypotheses and shares this information -- of course with lots of caveats about biases, not "scientifically collected", etc., with interested parties. They can then make what they wish out of it or use it as supportive information to get pilot funding for confirmatory research.

 

There are sources of bias and it's not perfect (and not interventional), but I think it would be an observational model from which we could learn a TON particularly about orphan diseases with highly motivated patients. Like having everyone's stories organized as data to be able to search patterns, etc.

 

I have actually already approached a well known clinical research survey company about setting this up as a model for orphan diseases, and have very strong interest. I think I also have funding through personal connections (royal family - not kidding), at least for a pilot, maybe more. But need to choose the pilot disease, finalize the initial data collection instrument with help of experts (have a general one drafted already), organize the medical and patient community around it, and figure out the ethics parts re. consultation with an IRB (e.g. code assignment for surveys to ensure anonymity). I actually think this could be such a promising approach to learning and sharing information about orphan diseases (and maybe ultimately other diseases) that I'm thinking about cutting back my consulting hours to pursue it more freely.

 

We should really talk.

 

 

 

Hi everyone,

 

 

It just dawned on me ...

 

Traditionally clinical research is done at some large institution, say the (John) Mayer Clinic (names changed to protect the innocent)

 

1. Dr. X, usually based on prior or often his own basic research formulates a hypothesis. Patients' ideas and observations don't figure much if at all in this step.

2. Dr. X attempts to "recruit" patients into his study, which must first pass an Institution Review Board, and many logistical hurdles. All sorts of safeguards are placed to ensure that the Mayer Clinic does not get sued.

3. Many months later, once those logistical hurdles are cleared, Dr. X opens his study

4. A year later, enough people have called Dr. X's office so that he has enough data to write a pilot grant to the NIH

5. Six months later, Dr. X finds out that his grant is just short of the threshold for funding

6. He submits another version, and prays - the Mayer Clinic is beginning to put pressure on Dr. X to bring in funding. After much white-knuckling, Dr. X finds his study is funded

7. Several years later Dr. X completes his study and finds out that his hypothesis is wrong

8. Dr. X is "non-renewed" (i.e. fired) by the Mayer clinic

 

Dr. Y, newly hired by the Mayer clinic to replace Dr. X, is inspired by Dr. X's work, and decides to embark on the same program as Dr. X

 

Dr. X, although disheartened, decides to try something new ....

 

There is something magical going on right here which might be a model for the way we do clinical research in the future:

 

1. Parents of patients (or with adults, patients themselves) tell their histories, make observations, discuss lab values

2. Interested doctors observe the proceedings

3. Doctors and parents formulate new hypotheses

4. Doctors set in motion ways to test the hypotheses by testing and treating patients (there's the rub)

5. Communal feedback on test results and treatments

6. Refine hypotheses and treatments to obtain improved diagnosis and treatment

 

What is the role of outside parties like the government and instiutions? really two-fold that I can see (but they are huge): ensure confidentiality and safety.

This has got to be incorporated in version 2.0 of the proces.

 

 

In the past 2 months alone, we've learned about mycoplasma and the p41 (anti-flagellin) connections by the new process. I'm sure there is more to come.

 

Can you appreciate how this rapid feedback iterative model VASTLY accelerates early clinical research? In the past, local doctors did this one patient at a time

 

Now, by using the "social networking" features of the internet, we can do it on large populations, i.e. statistically significant numbers.

 

The revolution will not be televised. But it will be broadcast on the internet and you are part of it!

 

Dr. T

[kimballot]

I would like to expand further on the "new" model. I think it would be great to take a formal survey approach - allowed any patients with verfied diagnoses to "opt in" to email mailing lists, and then each time a new question arises (by the scientific managers) or new data is created (by the patients), a survey expert formulats appropriate questions and patients can follow a link to enter their information into a well-designed survey/data collection instrument. Someone expert in data analysis then mines the data at appropriate intervals to look for patterns and generate/test hypotheses and shares this information -- of course with lots of caveats about biases, not "scientifically collected", etc., with interested parties. They can then make what they wish out of it or use it as supportive information to get pilot funding for confirmatory research.

 

There are sources of bias and it's not perfect (and not interventional), but I think it would be an observational model from which we could learn a TON particularly about orphan diseases with highly motivated patients. Like having everyone's stories organized as data to be able to search patterns, etc.

 

I have actually already approached a well known clinical research survey company about setting this up as a model for orphan diseases, and have very strong interest. I think I also have funding through personal connections (royal family - not kidding), at least for a pilot, maybe more. But need to choose the pilot disease, finalize the initial data collection instrument with help of experts (have a general one drafted already), organize the medical and patient community around it, and figure out the ethics parts re. consultation with an IRB (e.g. code assignment for surveys to ensure anonymity). I actually think this could be such a promising approach to learning and sharing information about orphan diseases (and maybe ultimately other diseases) that I'm thinking about cutting back my consulting hours to pursue it more freely.

 

We should really talk.

 

I am with you - mom to 2 pandas. Keep us posted...

[Dr_Rosario_Trifiletti]

I have a strong interest statistics and am an "R" and "S-plus" junkie for many years. The idea here would be exploratory data analysis where R and S-plus really shine. I'd love to talk to whomever is interested in doing this, other math/stat/computer junkies (PM). I've served as a statistician on a number of published papers but could use help from anyone with time, energy and interest.

 

I could propose some sort of data collection instrument.

 

I want to make this as "red-tape free" as possible.

 

Dr. T

[Chemar]

I do just want to remind everyone that certain things like data collection and research on this forum really does need to be cleared with Sheila Rogers FIRST please.

[Dr_Rosario_Trifiletti]

I do just want to remind everyone that certain things like data collection and research on this forum really does need to be cleared with Sheila Rogers FIRST please.

 

I do not plan to collect any data on this forum. Thank you for clarifying this.

 

Dr. Trifiletti

[melanie]

Hey Dr T

 

If you need any info we can do it on your site ?

 

Im in !!Anything you need count on us.

 

Melanie and Danny

[Buster]

Can you appreciate how this rapid feedback iterative model VASTLY accelerates early clinical research? In the past, local doctors did this one patient at a time

 

Dr T, I thoroughly agree. In fact, one of the reasons I put together the fact sheet and FAQ sheet was explicitly to accelerate the path of parents in coming up to speed as I wish I had had that information in March 2008.

 

What frustrated me so much about certain research by a very specific organization was that they were propogating their opinion in their title and not in their study. Titles like "PANDAS: Is It Important?" or "The PANDAS hypothesis: losing its bite?". Things like selecting patients who had 3+ years of continuous tics without remission and then making some statement about PANDAS. Only buried in their report can you find that they didn't repeat other people's experiments but decided to use rabbit brains (despite Husby's warning from 1977). Or that their children had zero fluctuations in their OCD severity (1.6 actually), when Swedo, Snider, etc explained fluctuations of 18pts. While I appreciate that their studies didn't confirm prior studies, I wish they had spent their time confirming/repeating the prior experiements rather than creating new experiments that could neither confirm or refute prior findings.

 

 

Best regards,

 

Buster

[faith]

Buster,

sorry, I am confused, .. are you saying the subsequent tests to replicate the Swedo study was done on rabbit brains and not real patients? sorry if I'm getting this wrong.... I thought Kurlan described on the Webinar how they picked controls that were matched ie. age, symptoms, etc......Is that the study you are referring to, or somehting else? .... rabbit brains? lol, how do rabbits have symptoms of ocd?