In 2017, researchers Los, Ramsey, Gottman-Bauman, and Ahmann published an exploratory study to attempt to evaluate the reliability of Diabetic Alert Dogs (DADs)1. One of the primary reasons for the study was that other authors’ previous research focused on perceived effectiveness of DADs, preventing medical professionals from providing information to patients regarding the physical benefits to those with diabetes and not just perceptions. 

Los and colleagues’ study incorporated eight DADs and their handlers over one week. Each handler had type 1 diabetes and wore the Dexcom continuous glucose monitor (CGM) for one week, blinded to the readings. Every time the subject’s dog alerted, the subject recorded their capillary blood glucose value (blood sugar reading from a meter) and their current symptoms. After the week, the researchers downloaded and analyzed subjects’ CGM data and DAD alert records. 

The authors concluded that DADs have a positive predictive value of only 12%, essentially only 12% of the dogs’ alerts actually occurred during hypoglycemia. This makes it appear that DADs are incredibly inaccurate and unreliable. However, the researchers failed to incorporate numerous critical aspects into their calculations, causing it to appear the dogs’ performances were very poor. The primary oversights are discussed below, demonstrating why this study was poorly designed and is not an accurate representation of the dogs’ reliabilities so additional research is needed.

1. Failure to account for hyperglycemia alerts. While the researchers’ goal was to focus on hypoglycemic events, they failed to consider that most DADs are trained to alert to hyperglycemic events as well. Every dog’s alert above 70mg/dL, even if during a hyperglycemic event, was considered a false alarm for the purpose of this study. The authors mention this was not anticipated when designing the study and should be studied further. However, this is not just something that “should be studied further”. This oversight caused an inaccurate number of false alarms because DADs which correctly alerted to hyperglycemic events throughout the week were considered to be alerting incorrectly, solely due to the authors’ failure to research DADs prior to designing the study. Furthermore, instead of reanalyzing the data according to the dogs’ training to alert to hyperglycemia, the authors acknowledged the limitation and released clearly inaccurate results.
2. Inconsistent hypoglycemia threshold. This study considered a DAD’s alert to be valid only if the subject’s glucose level was 70mg/dL or below at the time of an alert. However, a DAD’s trained criteria for hypoglycemia alerts varies based on the organization and client; some alert to ≤80mg/dL and others even alert to ≤100mg/dL. Setting a universal hypoglycemia threshold for the study without taking into consideration each dog’s training provides a falsely excessive number of false alarms, as almost all DADs are trained to alert to a threshold above 70mg/dL. The researchers even acknowledged that there were frequent alerts between 70-99mg/dL. However they only responded to this observation by calculating the percent of instances where the handler’s blood sugar would drop below the threshold within 20 minutes if it continued at that rate of change. While interesting, it is not nearly as relevant as considering the dog’s trained alert threshold. To take this variation into account appropriately, the researchers could have asked each individual what their dog’s trained threshold was and evaluated each dog according to its individual training. 
3. Inconsistent meters. Another aspect to consider in this study is the various glucose monitors used. Glucose meter accuracy varies greatly and four subjects utilized meters which did not pass the Diabetes Technology Societies (DTS) accuracy standards, so the dogs were being compared to a possibly inaccurate measurement. While there is no perfectly accurate glucose meter, to consider this the researchers could have provided subjects with the same brand of meter which did pass the DTS accuracy standards. 
4. Inaccurate CGMs. In addition to comparing the dogs alerts to inaccurate glucose meters, the researchers were also comparing the dogs alerts to CGMs which are known to be incorrect at times, which is often why individuals turn to DADs. The study claims the dogs alerted to only 36% of the lows according to the CGM data. But, the CGM is known to provide false alarms. In the Dexcom G5 FDA briefing committee documents2, when the CGM has a low threshold set to 70mg/dL for a pediatric patient, the true alert rate (PPV) is only 68% with the missed alert rate of 25%. When analyzing DAD missed alerts, the researchers were basing the data solely on a piece of technology which has a false alert rate of 32% meaning that the hypoglycemic events that the DADs “missed”, rely upon the CGM not having a false alarm at the time. The Dexcom is one of the most accurate CGMs on the market, so there is not a CGM that would have been preferable. However, the researchers could have incorporated an alert system which allowed the subjects to be made aware of CGM hypoglycemia alerts and prompted them to check their glucose with a meter, providing a more accurate result and confirming whether the DAD missed the alert or the CGM was giving a false alarm. 

While DADs are certainly far from perfect and this research asks an important question which needs to be answered, the study was not designed to properly analyze the reliability of Diabetic Alert Dogs. The authors failed to take into account numerous critical aspects that significantly altered the results of the DADs' reliabilities throughout the week of testing. 

1.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5505410/
2. https://www.fda.gov/media/98983/download

Filly is a 12 month old Comfort Retriever, in training as a Diabetic Alert Dog. Below are the commands that she knows at this point in time.

1. Sit (visual signal - j motion with hand)
2. Down (visual signal - flat hand lowering)
3. Stand (visual signal - two fingers pointing in front of dog)
4. Sit Stay (visual signal - palm moves towards dog's face)
5. Down Stay (visual signal - palm moves towards dog's face)
6. Stand Stay (visual signal - palm moves towards dog's face)
7. Come - come to front and sit
8. Heel - walk at left side
9. Here - come towards me
10. Let's go - walk with me (on right side or between legs, depending on previous command)
11. Watch Me - give eye contact
12. Touch - touch my hand with nose (visual signal - palm presented in front of dog, verbal required)
13. Leave It - do not touch or eat item
14. Under - lay under table, chair, or object (visual signal - pointing where dog should lie)
15. Close - sit at my left side, facing forward (visual signal - pointer finger drawing a 'U' demonstrating the dog's path)
16. Side - sit at my right side, facing forward (visual signal - pointer finger drawing a 'U' demonstrating the dog's path)
17. Between - sit between my legs facing forward (visual signal - point between legs)
18. Behind - lay between the back of my heels and counter or wall (visual signal - point behind, where dog's nose should end up)
19. Front - lay between the front of my feet and counter or wall (visual signal - point in front, where dog's nose should end up)
20. Chill - put chin on my foot (visual signal - two taps of foot)
21. Paws Up - put front paws on object that I point to
22. Get Up - jump onto object (visual signal - patting object with palm)
23. Get Dressed - put head through vest (visual signal - presentation of vest in front of her)
24. Get Your Vest - retrieve vest and bring to my hand (visual signal - point toward vest)
25. Get Your Leash - retrieve leash and bring to my hand (visual signal - point toward leash)
26. That's Trash - pick up item and put in trash can (visual signal - point towards item, then point toward trash)
27. May I have a yogurt? - open fridge, take yogurt, bring yogurt to me (visual signal - point toward fridge)
28. Get The Meter - retrieve meter and bring to my hand 
29. Hold - hold bringsel on collar
30. Bring It - bring object
31. Shut The Fridge - close the door to the fridge (visual signal - point toward fridge)
32. Shut It - close any door or cabinet (visual signal - point to object)
33. Get In - get in car 
34. Place - lay on place blanket for a long period of time (visual signal - point to place)
35. Wait - pause, typically at a curb or doorway while heeling
36. Break - release word (used for stays, when allowed to relieve herself, when allowed to play)
37. Off - jump off object (visual signal - point toward ground)
38. Drop - drop item or food in mouth (visual signal - opening of hand)
39. All Better - blood sugar is taken care of, puppy party is ending, stop alerting. (visual signal -flat hand moving from left to right in front of her face)
40. Powder Your Nose - relieve herself, regardless of location - on grass, mulch, concrete, grate. 

Filly turned 6 months old last week! We have been in Philadelphia for about 1.5 months and she has done great. She has adjusted without a problem and has become quite a city dog. We have a little tiny fenced backyard (if you can even call it that) that she hangs out in while I eat dinner or study. We also spend a lot of time on our front porch, where she has learned to be off lead. It's not fenced in but she now knows she isn't allowed to put even a paw off of it (it's pretty adorable when she plays with a toy and it falls off the porch so she lays down and stares pitifully at it). This is incredibly important because we have a busy street right in front of our house, but I love letting her be off leash, so this was our compromise. Since it's still warm out, I bring out a few toys for her and my homework and we will sit there for hours. She loves to people watch since we have so many people and dogs walking by our house. She's also getting quite the following of people who come up to the porch to say "Hi" to her.

I have not been going low very frequently, so she has not had much practice alerting unfortunately. She has been doing well alerting to samples though. She also seems to be alerting to drops and rises in my blood sugar, even though she never gets rewarded for it. I consider a fast drop or fast rise when my bg changes 10% or more in 10 minutes. The difficulty with fast drops and rises is that alerts to them get repetitive. If a diabetic gives himself/herself insulin, he or she will drop quickly. But they just gave themselves insulin so they know that, they don't need a dog to tell them. The majority of the time they dose correctly, so they will plateau at a healthy number in range and don't need an alert to it. Some organizations only train so their dogs alert to rises and drops and not to exact levels. While it's helpful sometimes, it also results in many unneeded alerts. Additionally, it requires a LOT more testing because without a CGM you can't tell a drop or rise is happening by just one finger prick, it requires you to recheck in 10 minutes every time the dog alerts in order to know if it's going up or down and by how much. 

All this is to say, that Filly is alerting to rises and drops which is great, but she still isn't getting rewarded for them. Another reason that she isn't getting rewarded for them is because I personally do not reward my dogs unless I can verify the alert - otherwise I will get false alerts. But to verify a fast rise or drop, I have to wait 10 minutes to recheck and see if it changed significantly. While I think my dogs are brilliant, they are still dogs and if I reward 10 minutes after an alert, they won't know what they are being rewarded for. I'm not sure if she will stop alerting to rises and drops if she continues to not be rewarded, or if she always will. If she always alerts to them it's not the end of the world as it is helpful. But she will need to be placed with someone who is able to check their bg very frequently and wants a dog that alerts to these things rather than just to 80 and below or 180 and above. Some people do want to know about fast rises and drops so it will just be a matter of finding the right person! 

Her alerting right now is to initially paw me, then to hold her bringsel and paw the air. We are working towards her alerting by holding the bringsel and pawing my leg rather than the air because sometimes she alerts but I can't tell because she's behind me pawing the air and I can't feel or see her. But she's doing well with holding the bringsel so I think she will get it. Sidenote, she is the most adorable thing ever when she alerts holding the bringsel. 

At 6 months this Crouton loves: morning snuggles, people rubbing her belly, doing zoomies around the house, throwing tennis balls in the air and pouncing on them, and snuggling (this is new!). Crouton does not love: pouring rain, boots, and having her teeth brushed. 

I just wanted to give a quick update as a follow to on the last post.

Filly definitely is live alerting! She gave her first live alert on July 9 and her next was on the 16th.  Her third live alert was today, July 19th. But so was her 4th and 5th live alert! Today she has alerted me 3 times to my personal low blood sugar! I am so proud of this girl!

The first live alert today was partially prompted. We've been on vacation the past few days and I thought to get her back on track (not that she was off track, but vacation is over so I need to work too ha!) I should set up a live alert. It was before lunch so I thought I would check myself, it turns out I was 72 with zero symptoms. I sat down next to her and dangled my hands just being near her. The combination of the scent and me having just checked my bg, caused Filly to take a sniff of me and alert. We had a party with lots of excitement even though it was prompted. I had to make lunch and my bg wouldn't come back up before then, so I encouraged her to realert, but then she rerealert about 3 more times after that, once she realized she kept getting cookies for realerting every few minutes! I think this may have hit home for her as she learned that as long as that scent is around, she should alert.

The next two live alerts today were unexpected and completely unprompted. For the second one, I was in the middle of working on place with her. So I would tell her "Place" and point to her bed, then give her a treat by tossing it off the bed and releasing her. Then we would repeat this a bunch of times. It's a pretty active exercise but I told her to place one time and she came in front of me, sat, and pawed my leg. I checked and was 74! Woohoo! 

The third live alert was not quite as crisp but still unexpected. I was pouring some food for her to use as training treats, when I noticed she was staring at me (not unexpected as I am pouring kibble!). But then I noticed my hand shaking slightly as I was pouring the kibble. I thought maybe she's trying to tell me something. I left the room and she followed, then laid down and continued to stare at me. Then she let out a piercing bark while still staring at me. I had stopped walking at this point, and just looked down at my phone, thinking if she was trying to alert I don't want to correct her, but if she's trying to alert I am not by any means going to respond to a bark! But after a few seconds of thinking, she sat up and pawed my leg. 77! Excellent demonstration of how to let the dogs think, because most likely now that she had to think through that process, she won't get in the habit of barking - she understands that barking gets nothing but pawing gets the cookies.

Excellent progress made by Miss Filly - I think she is really catching on and doing great for less than 4 months old!

Since I brought Filly home I've been working on both scent and obedience training with her. 3 days ago I started the step of scent training where I rub the saliva with low blood sugar on my skin and wait for her to paw me. With Filly I'm using a slightly different method than I did with my previous DADs (because a main reason why I enjoy DAD training is because I have the opportunity to experiment and tweak training methods all the time). The difference is that I've been rubbing the scent sample on the back of my knee and foot rather than on my hand. This is because I've noticed the previous DADs like to nudge my hand in order to confirm a low or high before actually performing the alert, but this requires them to sit up and reach my hand, and requires my hands to be at their level which isn't always possible as I'm often typing on my computer, taking notes, etc. The other DADs still alert very well, but I am curious whether it takes them a few more seconds to alert because they were essentially taught that the scent always comes from the hand. The back of my knee and feet will almost always be at a similar, accessible position for the dog to smell, and those are two places where people tend to sweat the most meaning that the scent of a live low or high will be fairly strong there.

Well today, I was able to see how Filly is putting this into practice! She did a tentative live alert for the very first time and when she pawed me the second time she sniffed the back of my knee prior to putting her paw on my leg. Good job Filly! 

Before the alert I was sitting on the couch on my phone, Filly was on her "Place" (Kuranda cot) right in front of the couch. She was staring at me and being restless in general, but most likely that is because she is a 3.5 month old puppy who gets really bored on her place. But because she is incredibly adorable when she stares at me with those big eyes, I leaned over to pet her. As I leaned over she sniffed my breath and then placed her paw on the couch right next to my leg. Skeptically, I said "Let's check" and we walked to my room where I began to check my bg. As I was checking it, she sniffed the back of my leg, and pawed me again. The meter read 80 mg/dL. I didn't really believe it as I hadn't washed my hands before checking so I retested and the meter read 82. With two readings that low, I figured that chances that it is somewhere below 82 was pretty likely, so we had a party! 

As she gains confidence I'll start to only reward her for alerts to 80 and below. But as this was the very first time alerting to a live low blood sugar, I wanted to build up her confidence and reward her for it. When dogs learn to live alert, they are typically much more tentative because they are thinking independently - a live low blood sugar is different than a sample because a live low blood sugar is 'moving' and doesn't include things like the cotton or plastic that the samples include.  

This does not mean that Filly knows how to live alert, will catch them all, or really even knows what she is doing. But it means she is likely starting to put the pieces together and think independently which is fantastic! There is still a chance this was all a coincidence but even if it was, this experience taught her that when this scent is present and she puts a paw near me, AMAZING things happen! Cookies come from all over, she gets a big bone, mom is super happy, and she gets a bunch of toys! So even if it was a fluke, she learned something from it. I will see if she repeats this within the next week or two and that will show me whether this was an intentional alert or just a coincidence. If it was intentional, a first live alert at 3.5 months old is earlier than my other dogs began, and I'm really excited to see what the future holds for this girl. 

Liberty's journey with me has ended, but her journey with her girl, Kylee, has just started. On May 10, Liberty was placed with Kylee as her Diabetic Alert Dog. I spent two days with Kylee, attending school with her, as well as practice and going about their life with them. This was to help transition both Liberty and the Dwinal family to their new lifestyle. My previous two DADs were placed at 9 and 10 months old. I wanted to experiment with Liberty, placing her a few months later when she was more mature. Liberty's behavior at 9 months and 14 months was pretty similar, however the transition to her new family went more smoothly. I can't know for sure if this is because she was more mature or because we had a gradual transition period (3 sleepovers with Kylee, many training sessions, etc.). But, I believe that a gradual transition period and a 12-14 month old dog is a better combination. Suzie, the dog I placed around 9 months, went through a 'depression' period when I placed her. She didn't have the opportunity to have a gradual transition, and she was very young, so the transition was hard on her. While it's not always possible, I would like to mimic Liberty's schedule to help ease the dog into their new life.

So far, Liberty's transition has been going well! She attends school with Kylee every day (which is a huge responsibility for both Kylee and Liberty!) and is learning her place in the family and what her new normal looks like. I'm incredibly proud of this new team. It's never perfectly smooth, but they are doing a fantastic job ensuring Liberty is the best she can be as well as giving her the love and family she needs.  Thank you to the Dwinal Family for taking care of my girly, I miss her more than words can say but she couldn't be in a better place, helping a better girl. 

Already Liberty is over 7 months old! She is still doing wonderfully with me at Penn - she doesn't have a single absence :) Everyone is used to her in class now, which makes me a little more relaxed. For the most part people have been very respectful of her. We occasionally have the 'driveby pettings' but because she's so short it's a little difficult for them to be discrete! I've been surprised how respectful other students are - they almost always ask to pet her first and usually are understanding when I have to ask them not to. The worst people are the staff - those that work security or cafeterias. Typically they are the ones that make the kissy noises and try to call her over and distract her. But she usually handles it very professionally and ignores them. 

Liberty is now a little over 6 months old and is 20.5 pounds. To say that she has done well here at Penn would be an understatement. We have been here about 1 month and Liberty has been an trooper through the transition. She has adjusted to city life very well - she's not phased by buses, ambulances, fire trucks, busy streets, other dogs, grates, weird steam coming out of said grates,  or riding the shuttle. She attends all my classes - so on our easy days we are gone for about 6.5 hours, some days are longer. She tucks under my chair in each class and sleeps through it. Her small size makes fitting her under my chair almost always doable. She does occasionally let out a groan of complaint in the middle of a class if she's bored though - this usually causes some laughter. 

Before this year I had never been to a dog park - I lived in a dog park practically. I still don't love dog parks because they make me nervous - all the strange dogs that have never met, running around with toys to guard and humans to take ownership of -  but I do realize that sometimes they do have benefits since I was worried about my dog not experiencing enough canine socialization. The first weekend I was in Philly Liberty and I went to the dog park. She likes other dogs so I thought this would be fun for her. She was terrified. The dog wouldn't go within 50 feet of the entrance for 30 minutes! Gradually we worked up to sitting next to the gate and watching dogs go in and out (During this time she was huddled between my legs looking like I was torturing her). But gradually she became more curious until we got to the point where we could go inside. Once inside, she reverted back to square 1 - huddled between my legs, terrified. After about another 20 minutes she began to wander to see the other people and sit between their legs - not quite the behavior I was looking for but a step in the right direction! Very gradually she investigated the other dogs, venturing out to see one then running back to my legs. Until a 6 month old Goldendoodle came in the park - these two hit it off and she finally played! Thankfully she has improved greatly and she now gets excited as soon as she knows we are walking to the park - she loves to play in the baby pool full of water that they have set up, and she plays with any dog that is smaller than her (she definitely likes to be the one doing the beating up, rude). I think that she has come to the realization that she is, in fact, a dog and not a human. Well, maybe she's just realized that dogs are fun to play with - she still thinks she's a human.

As always, we are practicing low blood sugar alerts frequently. She gets practice with live lows almost daily and about once or twice a week I practice with scent samples. Her precision has been improving - typically she alerts right at 80/81 or below. I've noticed she might be alerting to me dropping - a few times she's alerted and I've been in the 90's, I then tell her "We'll watch" and recheck in 5 minutes. When I recheck I'm often in the lower 80's or the 70's. She never gets rewarded unless I'm at 80 or below, but alerting to drops could be very useful to whoever she is placed with. 

We need to work on extreme lows as she is only familiar with lows in the 70's and upper 60's. A few days ago she gave me a very half-hearted, unsure alert. It consisted of pacing, stretching, yawning, and a little whining. These are often her 'pre-alerts', before she gives me the paw. I didn't feel low in the slightest and had just eaten and drank a soda recently. So, with this unsure alert I thought she was maybe bored and false alerting.  But I checked my blood and the meter read 46! This is very unusual so I rechecked and it read 67. Both times I used alcohol and my hands are clean, so I'm not sure which was accurate. I didn't start to feel low until my sugar was rising - and even then I didn't feel terrible. This is what makes training DADs difficult - each meter is different and even the same meter will give different readings. But, typically with enough tests I can get a sense of what the actual number is. And in this case, my sugar was definitely low and Liberty was right in that it was low. She got her puppy party and I drank some juice. 

A few of Liberty's favorite things: swimming, the dog park, people obsessing over her, ice cubes, hotdogs, morning snuggles, liver, and chasing squirrels. 

Some things Liberty does not like: being left alone, me not letting her chase the squirrels, walking past the dog park without going in, and being ignored. 

This little Munchkin is doing very well at just 6 months old! Currently, I'm thinking that Liberty would be best suited to be placed with an older teenager or a young professional. She needs someone that she can be completely loyal to and will be her "person" alone. She will also need to be placed with someone within a few hours drive of Philadelphia so that we can gradually transition her over multiple months. If you or someone you know could benefit from Liberty, feel free to contact me or fill out the DAD application (tab above).  

This past year I was in Philadelphia at the University of Pennsylvania for my freshman year of college. Just before moving to Philly I placed the second Diabetic Alert Dog I had trained.This meant that for the first time in 10 years I was dogless. To satisfy my canine obsession I worked at the PennVet Working Dog Center for 8-10 hours/week. I am so grateful for the opportunity to be involved in the training of their dogs. I discovered a new realm of working dogs, attended the Center's working dog conference, and worked in the ovarian cancer detection program. Although it was an incredible experience to be involved (and I'm not sure I could have survived without them), I still missed having my own dog that I could train. So, in October I began to inquire about having a Diabetic Alert Dog in training with me while at Penn.  Long story short, I asked multiple people, had numerous meetings, received countless rejections, but in a last ditch effort during the last month of school, God sent someone that finally approved my request to have a DAD in training! I was, and still am, ecstatic and incredibly grateful for the opportunity to continue my passion of training DADs during my sophomore year at Penn.

As she matures and her alerting improves, I will train her to alert using a bringsel. At some point I may try to teach her to alert with the bringsel hanging off her collar, but this will depend on her drive for the bringsel and if it works with her temperament. A lot of DAD training is discovering what methods work best for each dog. 

My plan is to place Liberty with a type 1 diabetic near Philadelphia so that beginning in early 2017 we can gradually transition her over to the new family. If you or someone you know is interested and in need of a Diabetic Alert Dog, Liberty will be available at the end of May to be placed with a family - when she is 14 months old.  Feel free to contact me through email, phone, or by filling out the DAD application (tab above) if you would like more information about her.  

I'm really excited for the upcoming year with Liberty at Penn, and I hope to update the blog at least once a month with her progress. For more frequent updates you can follow our Instagram account @libbysloving leashes or our Facebook page "Libby's Loving Leashes".