Sunday 25 September 2016

Episode 3: Battle of the CGM Systems: G5 vs MM640G, bolus and exercise adjustment with trend arrows


Welcome to the final instalment of Battle of CGM systems. I have already completed a detailed analysis and my thoughts of the four main CGM systems. This is a LONG blog so please feel free to jump just to the parts that interest you. This blog is includes:

  • 1.      A comparison of the very latest offerings:


a.      Dexcom G5 (including use of the Iphone APP) vs. MM640G with the newEnlite  inserter and new algorithm in Guardian Link as of August 2016.
                                                              i.      The only published abstract on the new MM640G algorithm to date is one showing a MARD of 9.1%. But that paper only measured MARD on day 3 for every 15 minutes for 4 hours after breakfast, not a comprehensive assessment by any stretch of the imagination. I am sure there will be more comprehensive research to come, but in the mean time the only way to know is to give it a go yourself.

b.     I wanted to see how they compared for accuracy, false alarms and missed hypos, and safety from bolusing from sensor glucose (SG) values.

c.       Did the new MM640G get to below 10%

d.      Using the Dexcom G5 App and sharing it with friends and family

e.      The new one touch Enlite inserter

f.        After all the analysis which CGM am I going to use moving forward?

  • 2.      How  I increase my bolus amounts if the CGM arrows are trending up, and how I reduce them if trending down. Which one of the popular three systems did I find best; the JRDF +/- 10/20%, Canadian TAAT protocol based on sensitivity factor, Dr Adolfsson’s +/- 10/20/30%



  • 3.      How I change the carbohydrate amount to manage exercise depending on the CGM trend arrows. Mixing exercise physiology theory, research on type 1 diabetes and exercise, and lots of clinical and personal experience. 


  • 4.      Some personal Diabetes philosophies. How developing a growth mindset helps learning and prevents frustration and burnout for me!


  • 5.      Some thoughts on future CGM technologies and diabetes devices. What would the ideal system have?



I have written this blog for a reader who has a good understanding of what type 1 diabetes is,  already knows what CGM is, has some understanding of the current products, and understands the different alarms you can use on CGM devices. If you currently do not have this understanding please link to the below articles, then this blog will be of much greater value:




Please remember whilst reading, this is a n=1 experiment and because the CGM devices worked in a certain way for me, does not mean they will work identically for you. Also some of the diabetes management strategies I use may be different to yours and what your diabetes team has taught you, so please question everything (I am my own diabetes expert, I am not yours), and consult your health care team before employing any of them. Disclaimers over, here are the questions I endeavoured to answer;

Dexcom G5 vs MM640G with new updated transmitter – which wins?

Before jumping knee deep into the comparison I want to bring your attention an amazing bit of work done by Dr. Kovatchev and his team in 2015. It is a heavy read so I have summarised the key findings here, but if you are really into CGM you MUST read this paper. They took the data of 56 patients with type 1 diabetes who had been on a pump and CGM for two years. They used innovative data analysis and methodologies to mimic what would happen in real life, for sensors accuracy ranging from 3 - 22% MARD. They looked specifically at;
  • ·         The safety of correction bolus decisions from the sensor glucose value
  • ·         Prevention hypos with low limit alarms
  • ·         The reduction of hyperglycaemia with high limit alarms
  • ·         Prevention hypos with predictive alarms
  • ·         All of the above were tested;

o   First without trend arrows to add further modification of therapy
o   Second with taking into account trend arrows to further modify therapy decisions to see if additional benefit can be gained

The main conclusions were:

  • ·         A MARD of 10% is needed to make safe and effective bolus, hypo prevention, and hyperglycemia reduction decisions.


  • ·         Although the safety and effectiveness does improve with MARD <10%, the improvement is very small and 10% is the critical threshold. The technical term is diminishing returns after 10%.


  • ·         For every 1% increase in MARD above 10% the errors increase in an "abrupt slope" manner.


  • ·         Using trend arrows to make further modification in bolus doses and hypo treatments leads to less hypoglycaemia and more time in target.


Put simply! To safely and effectively use CGM for all treatment decisions, the MARD need to be 10%!!! Every 1% higher incurs much greater risk of hypos and hypers, but every 1% lower does not make massive improvements in effectiveness. Also, using trend arrows to further modify decisions is beneficial and I will come to how I do that later in the blog.

It should be of no surprise therefore that there is only one CGM system, Dexcom G5 with FDA approval and CE mark to use the sensor glucose values for bolus decisions, treating hypos and complete non-adjunct therapy use. That is the Dexcom G5 because it has a MARD of  9% for adults and for 10% paediatrics. The Libre (11.3%) can be used for bolus decisions but not if the arrow is straight up or straight down, and hypos have to be confirmed with a finger stick. The Dexcom G4 (Animas Vibe 13-1 4%), Medtronic VEO and 640G (13-14%) have to have finger sticks for all bolus decisions, and confirmatory finger pricks for hypo treatment. The VEO and 640G can stop the insulin for up to two hours via Low Glucose Suspend and Smart Guard as the accuracy requirement is not as stringent for stopping insulin to prevent serious hypos, as it is for giving insulin which if done wrong can cause severe hypos.  This shows clearly the CE and FDA people know there stuff when it comes to giving clinical indications for medical device use.

I must state here that I am not promoting Dexcom per se and I am not against Medtronic & Animas per se. What I am for is CGM devices with a consistent 10% MARD that allow people with diabetes to make safe and effective decisions, to improve diabetes control and quality of life. I would love nothing more than for all CGM devices to have 10% MARD or better, so that no matter what device a person has they will get the benefit. I am hopeful this new algorithm in the MM640G will improve it to 10% and Medtronic go for CE mark for the same clinical indications as Dexcom G5. Then I can have the benefit of Smart Guard and making all my therapy decisions from it (read below to see if this was the case for me). I hope in the future;
·         The Medtronic 670G which will stop basal and give extra basal to stop high and lows has a MARD of 10%, as that will be game changer for overnight control. I worked for Medtronic for 4 years and know they are a great company that is heavily invest in research and development, innovative technologies and re-reimbursement. I very much look forward to banging the drum for their device that consistently hits a MARD of 10%.

·         I hope the ROCHE CGM device which is near has a MARD of 10%. Then with their Bolus Advisor technology (which in my opinion is hands down the best out there) will allow very effective post-meal sensor glucose corrections, because of its unique way calculating active insulin. The ROCHE Bolus Advisor calculates meal and glucose centric insulin separately, rather than lumping them both together which the other pumps do, leading to those pumps always having too  much insulin on-board. Maybe when that time comes I will do a blog on the pros and cons of the different bolus calculators (if time). Also the ROCHE bolus advisor allows you to see health events that can increase and decrease the bolus amounts by percentages ranging from +/- 0 - 50%. This would be perfect for what I discuss below in how the adjust bolus amounts according to trend arrows. So watch this space and ROCHE bring your A game to the CGM party.

·         I hope a company comes from nowhere such as Google or Sony  and brings out an amazing CGM that does things differently.  A little like Abbott almost have with Libre, as it does not require a finger stick calibration. But with no alarms and a MARD of 11.3%, which is 1.3% above the ideal benchmark, it does not quite hit the mark for me. I am sure if you had the option to calibrate it with the occasional finger prick (which I know you can do if you use an open source APP via NightScout, discussed in previous blog) then the MARD would get close to 10%. Then if you add the option of alarms by it being a transmitter rather than just recorder, you would be talking my kind of language.

I need to make those points above before delving into the analysis of my six day head to head comparison of the Dexcom G5 and MM640G with updated algorithm. See below the graphs for MARD %, false alarms and missed thresholds, and safety of bolusing from sensor glucose level.

Six day MARD Percentage head to head comparison of Dexcom G5 vs. MM640G (updated Guardian Link Algorithm as of August 2016)  14th - 19th September 2016




The Dexcom G5 was comfortably more accurate with an average MARD of 6%, and four out of six days with a MARD 8% or less! The MM640G with the new algorithm had an average MARD of 15%, which was at least consistent with no day was >21%.  The new algorithm MM640G performance was similar to the old algorithm for me. I tested it for another two weeks just to check, but the MARD remained at 13-15%. So it's clear for an accuracy point of view for me, the Dexcom G5 is much more accurate in terms of MARD %, and smashed the 10% barrier for safety and effectiveness. Let’s see how that safety and effectiveness translated for me translates into daily living with the sensor.



What impact does a MARD have on the number of false alarms and missed hypos?

I reviewed all of the downloaded information to look at how many false alerts at 4.4mmol/l and 9.0mmol/l I got with each system, and how many missed highs and hypos there were.

Number of missed highs, lows and false alarms; head to head comparison of Dexcom G5 vs. MM640G (updated Guardian Link Algorithm as of August 2016)
14th - 19th September 2016



Summary of the false alerts and missed highs and hypos:


·         The G5 shows why a MARD of <10% is so important. The G5 had virtually no false or missed high alarms, only one false alarms at 4.4mmol/l. There were two missed hypos but multiple prevented with the alarm set at 4.4mmol/l.

·         The MM640G performed similarly in all aspects to the G5 except, I had 6 false alarms in six days at 4.4mmol/l where my BG was >5.2mmol/l on all six occasions. This would have been very frustrating if I had the Smart Guard turned on, as that would have been six times where the insulin would have been stopped leading to high blood glucose level later on. This maybe something another person who has terrible issues with hypoglycaemia is happy to compromise with; some false suspends being a worthwhile cost for the protection of preventing severe hypoglycaemia via Smart Guard. For me however, the high blood glucose levels and frustration of false alarms is too big of a contrast compared to the accuracy of a <10% CGM system.



What happens if you bolus using the SG reading, does MARD % matter?

I looked to see if using the SG reading from the different devices at meal times to correct to a level of 5.6mmol/l would impact on my control. I used a similar methodology to Dr. Kovatchev for all of the bolus's in the six days, to find out what percentage of bolus's would end up in a specific target range:
·         within 1mmol/l of 5.6mmol/l (4.6 - 6.6mmol/l)
·         within 1.6mmol/l of 5.6mmol/l (4.0  - 7.2mmol/l)
·         within 2mmol/l of 5.6mmol/l (3.6 - 7.6mmol/l).

The below graph shows the results:
Percentage accuracy of correction bolus achieving target using Sensor Glucose readings; head comparison of Dexcom G5 vs. MM640G (updated Guardian Link Algorithm as of August 2016)
14th - 19th September 2016




This data again shows why having a CGM device with a MARD of 10% or less is needed to safely bolus from! Summary;

o   I would be happy to correct to 5.6mmol/l for G5. I would only have one glucose value land outside the range of 4.6 – 6.6 mmol/l every three days at the most, and never less than 3.6mmol/l.

o   I would not be happy to correct using the MM640G, as even if I corrected to 6.0mmol/l I would have one bolus every two days landing outside 4.0 – 8.0mmol/l

Combining the data from wearing all four CGM systems  for 14 days in the Grand Canyon (put in the link), with the six days comparing the Dexcom G5 with the new updated algorithm for the MM640G, leads me to one clear conclusion. I need a CGM device that achieves 10% MARD consistently to allow safe and effective therapy decisions from sensor glucose readings. The only device at the moment that meets that criteria is the Dexcom G5. Therefore I have changed my funding to Dexcom G5.

Two side notes I must quickly mention:

  • 1.      The Enlite sensor that goes with all Medtronic CGM devices has a new serter that is now just a one click. It is soooooo much easier to use and prevents any of those issues of pulling out sensors by nor re-pressing in the button. If only they had this when I worked for Medtronic my life would have been much easier! A great upgrade and if you use Enlite, you need one of these ASAP!


  • 2.      I used the G5 Iphone APP which allows me to;

a.      See my CGM on my phone and can be used independently from the receiver
b.      Share my CGM with up to 5 people who become your followers! Not a massive advantage to me but if I was a parent this would be invaluable. My Fiancée is following me and I am starting to understand why teenage kids get pissed off with their parents when they ask;
                                                              i.      "What happened there?"
                                                            ii.      "What did you eat to make it go that high"
                                                          iii.      "You were low for a few hours and did nothing about it?"
                                                           iv.      I will discuss this more in the personal philosophies and growth mind-set section.
c.       I linked to my Diasend account and APP which is sweet, I can review my CGM results easily via Diasend via their mobile APP
d.      If you are getting a Dexcom G5, get this APP, it's the Bomdiggy!


How I modify my diabetes therapy decisions based on the trend arrows from CGM

I discussed earlier in Dr. Kovatchev's research, if you modify your therapy decisions based on trend arrows from cgm, you get less hypos and fewer highs. The big question is how do you make these adjustments? Is it just pure trial and error, or are their guides out there? What I detail below is my method which is based on:


  • ·         Exercise physiology theory & research from my academic background and love of knowing how things work.


  • ·         Clinical experience from work, because what sounds good on research paper does not always translate practically. I see this as my most effective skill, turning the research into practical tools that can be used by people with type 1 diabetes.


  • ·         A large dose of trial and error. As I discuss more in my personal philosophy on diabetes, I am an APE! That is a person who takes Action, has Persistence and loves Experimentation. If it does not work the first time it does not mean it will not work, it just requires a different approach and another go!


Using these techniques with the Dexcom G5 sensor I am able to mainly keep my glucose level between 3.5 - 9.0mmol/l,  with 1-2 mild hypos per week that I can manage myself. This is evidenced by my BG readings and sensor graphs below. I do not put that there to show off, but to show how being an APE and using trend arrows you can get good control.







The first job was is to find out what all the trend arrows mean scientifically, and convert them to real life speak. The chart below consolidates the findings from my first job.



Once I had this information I could then go about trialing the strategies used in research papers and adapt them for myself.

Bolus adjustments based on trend arrow direction

There are three main approaches to adjusting bolus amount based on rate of change of sensor glucose arrows;

  • ·         CHEO group TAAT protocol - the most conservative
  • ·         The JRDF +/- 10/20%  
  • ·         Dr. Adolfsson +/- 10/20/30%   - the most aggressive


I have summarised what that the three methods look like in the graphic below.


You will see I have only included the Dexcom G5 for all arrows, and  the Libre only up to angled arrows. I have done that for a few reasons;
  • ·         Personal experience of adding 10-30% of insulin based on sensor trend arrows from a CGM system with 13-20% MARD has not been good. A higher MARD % really magnifies the error and has put me hypo many times. I have tried all these methods over the last two years whilst on the VEO and MM640G, and found big issues with hypos so I abandoned it and continued to use finger pricks for bolus decisions.


  • ·         When using the three methods with the Dexcom G5 I did not have the above issues,  the MARD of 10% or less made all the difference. The Libre also worked much more consistently, especially when glucose was moving slowly, but I had a few issues when moving rapidly.


  • ·         The graphic follows the CE mark and FDA approval, and I as much as anyone like to challenge dogmatic rules, but in the case of 10% MARD being the holy grail, I believe in that rule.



The only method that should need explaining is the TAAT protocol which is actually the simplest to use and most conservative. It uses the Medtroic VEO arrow system. I have a sensitivity factor of 3mmol/l, therefore the only two numbers I need to know are 0.5 and 1.0. If my sensor glucose value was 6.0mmol/l at a meal time and my bolus came to 7 units, the amounts I would give are as follows:
  • One arrow up = 7.5 units
  • Two arrows up = 8.0 units
  • One arrow down = 6.5 units
  • two arrows down - 6.0 units 

·
The TAAT was developed for using the VEO and does not take into account the new two arrows of Dexcom G5, so I added that column for me below. So the only numbers I needed were 0.5, 1.0 and 1.5.


I have below given you some examples of the difference that would mean in bolus amounts at meal times for different carb amounts and sensor glucose values. I have mixed the examples up to mimic at different insulin sensitivity levels and ages.







My assessment of the different methods is:
  • 1.      When the carbohydrate amount and sensor glucose value is not too drastic, all three methods come out with similar bolus modifications.


  • 2.      When the values are much larger the 20 - 30% changes are a lot more aggressive than the TAAT protocol.


  • 3.      To me this approach is just a natural progression when using a CGM system with 10% MARD. It is important to consider individual circumstances, some key factors I considered were:


a.      I am an APE, so I like to try the most extreme thing first and work back from there. I really like Dr. Adolfssons method and it has worked well for me. However, I do run very tight control and very rarely have two arrows so realistically very seldom use the 30%. if I was a young child I may not be so bold.

b.      How good is my math's on the fly? Working out 10,20,30% can be tricky if you are not a math's whizz. I found it easy enough (still made some mistakes), but I know from a lot of clinical practice that simplicity is the key for people to use things safely and effectively. That is where the TAAT protocol comes in, it is by far the easiest and most conservative. So if I was starting out I would definitely try that first.
                                                              i.      I hope in the future bolus calculators have the ROCHE health events that allow adjustment by +/- 0-50%. I get excited thinking that ROCHE may included this in there next CGM/pump device, where the calculator would take not of the trend arrows and suggest a percentage change!

                                                            ii.      Who else will be making the bolus adjustments decisions. I make all my own so I trust myself to do it right. If I had a child with type 1 diabetes at school, I would probably be more keen to start simply with he TAAT. I think teachers, grandparents, nannies and the like would appreciate the TAAT to start with.

                                                          iii.      Am I willing to put in the time and trial them?  If you just want to try something and not monitor closely, then maybe start with the TAAT.

What is the research behind all this I hear you ask? Well not a massive amount, but I have summarised the two key papers below, and they show promise;

1.      I have already discussed the Kotachev paper which showed that further adjustment to bolus amounts using trend arrows (the JRDF +/- 10/20% method) improved time in target and reduced hypoglycaemia. BUT ONLY IF MARD 10%!!!!!! The higher the MARD, the greater the bolus errors, and bolusing off CGM trend arrows that have a MARD well above 10% may cause more harm than good.

2.      The Canadian CHEO group put their TAAT protocol to the test against the JDRF method. They recruited twenty type 1 adolescent children and got them for a week each to; week 1 ignore the arrows; week 2 use the JRDF, and week 3 use the TAAT protocol. They looked at the time in target, in hyperglycaemia (>10mmol/l), and in hypoglycaemia (<4.0mmol/l). The two graphs below are from their abstract as the full paper is soon to be released.


You can see from the results that using trend arrows to modify bolus amounts improves time in target and reduces hypos, especially when the glucose level is in target to begin with (the top graph). So reduced hypos and more time in target, why would you not use the trend arrows?


The above graphic shows what I suspected, the JRDF protocol does on average lead to larger modifications which are potentially more beneficial, BUT the error rate was 17% compared to 1% for the TAAT. No surprise that more parents and children said they will use the TAAT moving forward. So make up your own mind, are you better with option one or option 2:
  • 1.      The TAAT protocol which is easy to use, ensures consistency and should be able to be used by all people caring for someone with type 1 diabetes.


  • 2.      The JRDF of Dr. Adolfsson percentage approach which potentially is more effective in terms of time in target, but comes at the risk of increasing errors.


I am sure you have guessed I am with Dr. Adolfsson, but that is because I am an APE who is independently managing my diabetes with reasonable math's skills. If I had a child with type 1 who goes to school, child care and sports coaching, then maybe I would lean to the TAAT, or more likely develop an APP that does it for them!

Exercise carbohydrate adjustments based on trend arrows

The trend arrows have been a revelation in the management of exercise for me. In the early days a blood glucose level of 6.0mmol/l before an hour doing exercise would mean I drank 30g of carbohydrate of Lucozade and get stuck in. What happened to my glucose level during and after would be a lottery, because at that time I was not aware of the below factors influencing:
1.      Active insulin from a previous bolus
2.      Duration in minutes
3.      Intensity: medium, high, very high
4.      Type of activity: Endurance, Intermittent, Short-sharp
5.      Weight: carbs per kilogram required
6.      Where is my glucose level trending: up, down, stable
7.      How often can I check my glucose level and consume carbohydrate accordingly?

To cut a 10 year long story very short I got my APE on and read, experimented and finally made a tool for myself, to take into account the above variables. This tool allows me a good place to start for new activities and planning trips such as the Grand Canyon. I have put a few examples below of what it looks like.


The plan above is for a flat days trekking where intensity is medium and activity type is endurance. Whereas on the same trip the climbing days were high intensity but still endurance and you can see the carbohydrate amounts for every 30 minutes increase.



Below are three plans for the typical activities I do; Weights session, HIIT session and Yoga. All are for 60 minutes but you can see the amount of CHO needed is very different. I could do it for 30 minutes then check but in reality once I get started I am not stopping! It worked for me in the Grand Canyon and continues to work for me every day.









I have shared this with quite a few Diabetes Teams in the UK and their response has been positive. There are few teams considering trialling it with their customers. I would love to share it with you directly but that would be stretching the boundaries, even for me. If you were desperate to give it a try I would ask your team to contact me and I can pass it to them. They may take a look and think it is a load of "dog poo" and they may well be right. I do like to break rules, but not those rules that lead to me losing my job!

My Diabetes Philosophy - The long evolution to a Growth Mind-Set

If you despise psychology and are not interested in my journey please feel free to skip to the future technologies section.

This part has been influenced by many sources including Seneca's writing on stoicism, positive psychology and the Tim Ferris Podcast to name but a few. One book however trumps then all, Mindset by Professor Carol Dwek. In my opinion an essential read for everybody, especially if you have a long-term condition.

I was diagnosed at the age of twenty five with type 1 diabetes, and back then I was a hard charged personality, where everything was pass or fail with no middle ground. For example;
  • ·         "I must be a professional football player and anything less is not good enough." I played for Everton until 16, got released as I was not as good as the others, in my eyes failure, so never played again, even for fun!


  • ·         "I must get 1st Class degree or it's a failure", consequently every second possible went into studying, 1st Class achieved through fear of failure but with social life consequences.



The overarching theme was I had to be the best at everything because I was a natural, and if I was not I would see it as a personal failing that lead to a negative view of myself. I could not keep the perceived failing to the specific task such as football. With this type of thinking my initial diabetes success parameters and consequently my personal success parameters were;

·         Every blood glucose should be between 3.5 - 9.0mmol/l
·         I must count every gram of carbohydrate to the gram, and later every fat and protein to the gram
·         I must check my glucose 10-15 times per day to be sure it's in target
·         If the glucose was not in target it was something I had done wrong
·         I must have the best HbA1c going
·         I must be prophetic and tell everyone else with type 1 diabetes how to manage their diabetes

As you can imagine holding myself accountable to the above standards lead to an inevitable conclusion, failure, and a lot of it that began to negatively erode my self-worth. So what do you do when that happens? Some people get depressed, some anxious, some withdrawn, but for me it was time the PARTY! Forget the tight control of diabetes, hit the party circuit and stick my head in the sand Ostrich style!

In steps Carol Dweck's  Mindset into my life. I had the book recommended to me by a mate who said it was one of those books that just hit the right note with him. The basic premise is that if you have been brought up getting praised for the outcome of success, and this gets associated with you having innate ability, god given talent, natural intelligence, you see things as you can either do them or not, based on what natural gifts you are born with. This would lead to a person having a Fixed-Mindset. As you can guess I had a lot of sports coaches, teachers etc who would say "John, you are such a natural talent, I bet you hardly have to practice", "John you are good at cricket, football, basketball, I wish I was born with that talent" and on. This is the basic reinforcement that develops a fixed mind-set. This is all well and good until you are not the best anymore, just like at Everton where the other boys were in my view "more gifted" than me, so instead of increasing effort, I gave up for good!

If you have teachers, mentors, friends, sports coaches that praise on the effort you put into achieve success, you associate success with the effort. Therefore when you get set backs or you are no longer the best, it's not because of a lack of natural ability of intelligence, it's just time to work a little harder or try a different approach.  You will certainly not be giving up because someone is better than you, as you will believe they have just worked a bit harder up to now, and that it's time to put a shift in.

So from this I developed my own little mantra, be the APE. Approach everything that is worth doing with action, persistence and experimentation. If I do not experience success it just means I need to change my approach and try something different. If I do fail it is not because I am a failure, its because the approach I tried was not right, and it just needs a little tweak, I am still a good person! So every time I catch myself dropping back to old habits of a fixed mind-set, I remind myself I am an APE and that usually gets me back on track!

Without becoming an APE there is no way I would have spent the time developing the above tools and trying all the different options. I would not have worn all the CGM devices in the Grand Canyon and read all the research. The best part is that by taking an APE approach I loved every minute of it, as it was all a chance to learn, grow and improve.

So a little plea from me to you if you have type 1 diabetes, or you are a parent of a child with type 1 diabetes. Look to foster a growth mind-set by praising the APE and making sure that failures and set-back are the springboard to try something new to learn, grow and improve.

Future technologies

If someone gave me a magic wand and I could combine the best of technology out there, what would I come up with?

  • 1.      A patch pump that has a catheter from insulin and a sensor in one - with a separate handset or linked APP to your phone


  • 2.      It would be a Dexcom G5 sensor for best accuracy


  • 3.     The APP/handset would have the ROCHE Bolus Advisor technology for the best post-prandial control. Also with health events to allow all the exercise and bolus adjustments


  • 4.      It would have the Medtronic 640 Smart Guard and the upcoming 670 increase basal to ensure perfect overnight blood glucose levels


  • 5.      The APP/Handset would have a food daatabase that you would put the portion size of the find in and it would work out the carbs, fat and protein and use the upcoming algorithms that account for their effect on after meal blood glucose levels 


The wrap!

Well this was a beast of an article and I hope it has given you some value. I certainly enjoyed writing it and I am grateful you took the time to read it. If you feel it could help someone, please pass it on.

This is the last blog for a while, or until some company comes up with something outstanding and I manage to get my hands on it! Until then I have some babies to make and a wedding to plan!

Cheers


John