August 30, 2012

The FDA Comment that Got Away... or Maybe Not

I like to see the DOC represented on FDA dockets that impact living with diabetes with comments. I think living with and self managing diabetes gives  patients insights into care that only patients know. So when a FDA docket get by me without my seeing it I am disappointed at the lost opportunity. One such lost opportunity was the Draft Guidance on Mobile Medical Applications. The docket was open this time last year.

I had a very pleasant chat with one of the FDA media specialist and she said that, yes the docket is closed but comments can still be made. Obviously comments will not have the same value to a close docket. So this is what I will send, late. Very late.

Consideration of the Regulation of Mobile Apps and Diabetes Care

The FDA has proposed guidance for the regulation of mobile medical apps. The draft is significant to people living with diabetes, in no small part this is because the FDA makes numerous specific references in the draft to diabetes care. Clearly then the agency sees diabetes as an area where mobile medical applications bring vale. The FDA Draft specifically speaks to diabetes care and blood glucose testing with strips, the analysis of information and an app controlling an insulin pump. 

Areas of the FDA draft where I think there may be a concern for people with insulin dependent diabetes.
  • The draft does not create a regulatory path for applications that process information from all the different devices people with diabetes use daily.
  • The draft reaches beyond mobile by suggesting regulation cloud based applications, if they are viewable on mobile devices. 
  • The draft is not clear about the distinctions between logging and analyzing data, particularly from the perspective of diabetes logging. 
  • Good apps are needed to improve quality of live. Failure to regulate in a timely fashion is a public health risk.

Multiple Devices
Effective care for people living with diabetes, particularly type 1 diabetes (YT1D), balances a variety of diverse things. Blood sugar is influenced by food, insulin and activity. Beyond the absolute level of blood sugar a trend, up or down, and the rate of that change also has an impact on diabetes management decisions. Different devices help people assess each of these different aspects of living with diabetes. The vast majority of the time T1D patients take self management actions independently based on data from multiple devices based on the training they have received from their health professionals. 

The draft says, “The FDA plans to address in a separate issuance mobile medical app intended to analyze, process, or interpret medical device data (electronically collected or manually entered) from more than one medical device.” This is particularly significant for the diabetes community. As mentioned above people with diabetes use multiple devices routinely throughout everyday to analyze how to manage diabetes. We use blood glucose meters, continuous glucose sensors (CGM), insulin pumps and carbohydrate food indexes. So while this draft makes specific examples of individual diabetes care apps, the integration of information from the multiple devices mentioned in the draft into one app is specifically excluded from the guidance. 

I see the next significant increase in the improvement in the quality of life for people living with diabetes coming from this integration of diabetes data with mobile devices. I see an increased amount of time in target rages and decreases variability in blood glucose through the integration of diabetes data from multiple devices. The fact is each device is used for only a part of the balancing act that is life with type 1 diabetes. Patients continually juggle blood sugar levels, the direction of change in blood sugar, insulin carbs and activity. An effective diabetes management app will consider the same multiple sources of information that people balance now. This draft specifically precludes applications that interpret data from more than one device. That exclusion maintains a status quo of regulatory uncertainty in the development and commercialization of tools for better diabetes self management.  

Artificial pancreas projects are being developed at a number of research facilities and health care firms. They will rely on data from CGM to control insulin pumps. The advanced logic of these devices may well be processed in mobile devices. These systems are in development and their commercialization will depend on the Agency being prepared to consider them as apps that interact with multiple devices. Yes the Agency is developing AP guidance. 

Low Glucose Suspend, a stepping stone to artificial pancreas, is available in virtually every market but the Untied States. However industry could not bring it froward for regulatory consideration in the USA because the FDA was slow to develop LGS guidance to evaluate it. The Agency should avoid following that unfortunate precedent, specifically where a lack of multiple device mobile medical app guidance inhibits AP.

Cloud
Cloud based diabetes software offers great potential for housing and considering data collected from multiple diabetes devices. The draft attempts to define what constitutes a regulated mobile medical application. In doing so they include applications that, “can be executed (run) on a mobile platform, or a web-based platform, or a web-based application that is tailored to a mobile platform but is executed on a server.”

This seems to allow regulation of cloud based computing platforms as mobile devices because the user has mobile access to output from that cloud system. That seems to be an excessively broad inclusion of essentially all cloud based systems as mobile. However in another section of the draft the FDA notes that it was previously determined that, “it would be impractical to prepare an overreaching software policy to address all the issues related to the regulation of all medical devices containing software.” It seems that reasoning should be applied to the cloud as a software containing entity. The attempt to regulate cloud based software, that can be seen on a mobile device, is also impractical. 

Log
The draft makes the distinction that apps that log general health and wellness information not intended for treatment are not regulated. In diabetes there is not a distinction of logging general health and for treatment. Further the distinction between logging and analyzing logged information is not clear. Many people with diabetes log blood glucose information as there is no manipulation of the values, the same values are simply visually represented. The display of that information as a table or a graphic such as chart should viewed similarly as displaying log information. Charts are simply a means of viewing logged numbers. A chart is not an analysis. This is particularly true when the device that an app connects with, such as a blood glucose meter, may already provides charts of the logged data.

Some devices are coming to market that do preform rudimentary analysis. For example some blood glucose meters now identifying blood glucose trends. The display of that information to an app is simply a second screen.  As such that second screen should be viewed as an accessory to the original approved device and not an independent regulated device. 

Failure to Regulate
The proposed guidance states, “The FDA believes that this subset of mobile apps pose the same or similar potential risk to the public health as currently regulated devices if they fail to function as intended.” It is significant the Agency considers the risk of a lack of devices. The agency should expand on that  logic. Yes, there is a risk if devices are not available and that lack of a devices is a potential public health risk. 

The FDA should consider the risk of a lack of devices stemming from regulatory uncertainty as urgency to develop the regulatory frame work for devices that can improve the quality of life and even possibly save lives.

Therefore I see the lack of diabetes apps as a public health issue.  So too is the lack of guidance for integrating multiple devices in an app. Consider for example that the diabetes community has mourned numerous children and teens who have died in the recent years ‘dead in bed.” Most likely due to extended over night hypoglycemia. These deaths should be considered as the outcome of the Agency failing to act to bring tools like low glucose suspend into the regulatory structure here in the United states. In that context the Agency should consider the lack of guidance for apps that “analyze, process, or interpret medical device data (electronically collected or manually entered) from more than one medical device” a public health risk. 



Possible form letters form this. 
Warning My Spelling May Vary!


The Honorable (You Congressperson or Senator here)
(Their address)
Washington, DC Zip

Dear (You Congressperson or Senator here),

The FDA is currently in the process of finalizing guidance for industry on how they will regulate Mobile Medical Applications (see http://www.fda.gov/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm263280.htm.) I (my child) live(s) with diabetes. 

Effective care for people living with diabetes, particularly type 1 diabetes (YT1D), balances a variety of diverse things. Blood sugar is influenced by food, insulin and activity. Beyond the absolute level of blood sugar a trend, up or down, and the rate of that change also has an impact on diabetes management decisions. Different devices help people assess each of these different aspects of living with diabetes. The vast majority of the time T1D patients take self management actions independently based on data from multiple devices based on the training they have received from their health professionals. 

The draft says, “The FDA plans to address in a separate issuance mobile medical app intended to analyze, process, or interpret medical device data (electronically collected or manually entered) from more than one medical device.” This is particularly significant for the diabetes community. As mentioned above people with diabetes use multiple devices routinely throughout everyday to analyze how to manage diabetes. We use blood glucose meters, continuous glucose sensors (CGM), insulin pumps and carbohydrate food indexes. So while this draft makes specific examples of individual diabetes care apps, the integration of information from the multiple devices mentioned in the draft into one app is specifically excluded from the guidance. 

I see the next significant increase in the improvement in the quality of life for people living with diabetes coming from this integration of diabetes data with mobile devices. I see an increased amount of time in target rages and decreases variability in blood glucose through the integration of diabetes data from multiple devices. The fact is each device is used for only a part of the balancing act that is life with type 1 diabetes. Patients continually juggle blood sugar levels, the direction of change in blood sugar, insulin carbs and activity. An effective diabetes management app will consider the same multiple sources of information that people balance now. This draft specifically precludes applications that interpret data from more than one device. That exclusion maintains a status quo of regulatory uncertainty in the development and commercialization of tools for better diabetes self management.  

Please connect the FDA and urge them to develop guidance for mobile medical applications that integration and manage the multiple devices diabetes patients use to self manage the condition. 


The Honorable (You Congressperson or Senator here)
(Their address)
Washington, DC Zip

Dear (You Congressperson or Senator here),

The FDA is currently in the process of finalizing guidance for industry on how they will regulate Mobile Medical Applications (see http://www.fda.gov/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm263280.htm.) I (my child) live(s) with diabetes. 

Cloud based diabetes software offers great potential for housing and considering data collected from multiple diabetes devices. The draft attempts to define what constitutes a regulated mobile medical application. In doing so they include applications that, “can be executed (run) on a mobile platform, or a web-based platform, or a web-based application that is tailored to a mobile platform but is executed on a server.”

This seems to allow regulation of cloud based computing platforms as mobile devices because the user has mobile access to output from that cloud system. That seems to be an excessively broad inclusion of essentially all cloud based systems as mobile. However in another section of the draft the FDA notes that it was previously determined that, “it would be impractical to prepare an overreaching software policy to address all the issues related to the regulation of all medical devices containing software.” It seems that reasoning should be applied to the cloud as a software containing entity. The attempt to regulate cloud based software, that can be seen on a mobile device, is also impractical. 

I support practical regulation of medical devices. I worry that impractical regulation may delay life saving advances. I particularly a concerned that innovation created by American firms are delayed from entering the market here due to impractical regulatory rules. The diabetes community has seen such delay. Please contact the FDA and express my appreciation for practical regulation of mobile medical applications but not what the Agency itself has called overreaching. 

6 comments:

Scott S said...

You know what my thoughts are: it's our legal right as American Citizens to comment on any guidance (or draft guidance) and anything the FDA puts out at any time, legally. So we can still do this, but I've been overwhelmed with work and other stuff that I didn't have time to put into this, but I concur. Now, how do we want to handle this? A Citizen's Petition, a formal comment on closed guidance documents? A number of possibilities remain open to us for this!

Bennet said...

Nice call for action Scott.

I added two draft form letters. People who are competent at spelling, punctuation and the niceties of the written language are more than welcome to point out my mistakes. I am sure there are a lot of them.

BD

Bernard said...

Thanks Bennet for a well thought out comment. This is such a challenging issue. As folks living with diabetes we know that insulin pumps and glucose meters are regulated, but it's still way too easy to kill yourself by accidentally misunderstanding the information presented, or having incorrect settings. In a private email to someone at the FDA I proposed that lack of unified access to data could be seen as a danger, unfortunately that currently isn't taken as a valid argument. We do need well thought out applications, so how do we balance safety with timeliness?

Bennet said...

Your are welcome Bernard.

It strikes me as totally illogical that easily killing yourself by confusion between devices, misreading them or setting them up wrong is not seen as a valid concern.

I suggest that part of the issues is to see that data from multiple devices is all part of care of one individual. That standardizing information has the potential reduce confusion and misreading and so improve safety. In addition standardized data should reduce the effort needed by each firm to create their own data protocols and speed up development in the long run.

kevinlmcmahon said...

In my experience in diabetes technology and as a former member of the CDMA Steering Committee, standards have a funny way of slowing innovation and driving the industry toward sub-optimal implementations. Can standards be helpful? Yes. But beware of an FDA that regulates based on a published standard vs. an approach that encourages innovation by those who drive forward ahead of the herd albeit with safety mitigation practices squarely in place (ie IRB oversight and adherence to methods published in peer reviewed clinical research).

Bennet said...

Thanks Kevin. Appreciate the insight.