Human versus Dog Infection rate Discordance

Medical topics with questions, information and discussion related to Lyme disease and other tick-borne diseases.
Post Reply
User avatar
LHCTom
Posts: 341
Joined: Mon 22 Oct 2012 4:18

Human versus Dog Infection rate Discordance

Post by LHCTom » Fri 2 Oct 2015 19:17

There exists a suspicious disparity or discordance between veterinarian reported Lyme disease in dogs and that of humans between Sonoma County California and New London Connecticut. New London is near the worst Lyme rates in the US and near where Lyme was first discovered in the US in 1975. Since dogs are good surrogates for their owners, one would expect dog infection rates to roughly match human rates by population. The hypothesis is that the dog infection rate should remain similar by population as the human rate while absolute numbers should vary by factors like dog behavior and inability to avoid and have ticks removed due to hair.

The dog infection rate of infections is similar between Sonoma and New London while human infection rates ratios are disparate by about 150 times by population. This suggests human doctors and tests used in Sonoma are different and inferior to similar doctors and testing in New London CT. Dogs are primarily tested by C6 ELISA which is more strain and species independent test while humans are typically tested by a single strain ELISA followed by a single strain Western Blot. I would suspect the Sonoma doctors are biased and ill informed and the CDC 2T test based on one strain is less effective in California due to 2500 miles and Rocky Mountain barrier caused strain diversity. Borrelia miyamotoi (Bm) was detected in 1/2 as many ticks as Borrelia burgdorferi for the last decade in Northern CA by the CDPH microbiologists doing tick testing. Bm is not as reliably detected by the CDC 2T than the C6 as is Borrelia bissettii also common in CA. The C6 dog testing may be superior.

C6 in humans

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4052829/

http://cid.oxfordjournals.org/content/47/2/196.full

C6 in dogs

https://www.idexx.com/pdf/en_us/smallan ... isease.pdf

https://www.idexx.com/files/small-anima ... -guide.pdf

The CDC acknowledged in 2013 that the human reported rates of Lyme disease nationally were approximately 1/10th of actual rates. Lyme Connecticut is where Lyme disease was first identified in the US in 1975 and CT continues to have among highest rates of the disease and tick infection rates. New London County, Connecticut had 1395 cases of human Lyme disease reported between 2011 and 2014. During the same period, Sonoma County California had only 33 reported cases. The lower rate of human Lyme disease in Connecticut is typically explained by the higher tick infection rates. But that should equally affect dog infection rates. It doesn’t and by a substantial amount of more than 2 orders of magnitude or 100-150 times. This dwarfs the CDC 10x human reported versus actual rates. Something is not right. The rate of infection in Northern CA is being underestimated by a significant amount. This in turn results in a self perpetuating feedback that causes doctors to believe Sonoma is low risk.

= Big Red ignored FLAG!

http://www.cdc.gov/media/releases/2013/ ... sease.html

The dog infection rates between Sonoma, CA and New London, CA counties have an enormous discrepancy between human infection rates when comparing Veterinary versus Doctor reporting. This cannot be right. The New London population is about 300,000 people while Sonoma is close to 500,000 or 1.7 times larger. The ratio between human populations and dogs in London County is roughly 4 people to one dog. The Sonoma County ratio between the human population and dog population is also about 4 people to one dog. So one would expect the Lyme infection in dogs in New London County and Sonoma County would have a ratio of about 1.7 since the ratio of dogs is near identical. The Companion Animal Parasite Council collects veterinary surveillance on various pet parasites including Lyme disease by US state and county.

http://www.capcvet.org/parasite-prevalence-maps/

The number of dog infections in 2015 in New London CT is 1167 and Sonoma CA is 1269 based on CAPC surveilance. That means New London had slightly less reported dog cases of Lyme than Sonoma County while New London is considered highly endemic and near Lyme central. But the New London human and dog population is higher. New London had an approximate dog population of 300,000/4 = 75,000 dogs which translates to 75,000/1167 = 1556 infections per 100,000 per year in 2015. Sonoma had an approximate dog population of 500,000/4 = 125,000 dogs which translates to 125,000/1269 = 1015 infections per 100,000 per year in 2015. Even though reported numbers fluctuate by year, the large discordance exists for any period. So the infection rate per 100,000 in dogs is only a ratio of 1556/1015 = 1.5 times. The ratio of human infections should be similar while the absolute numbers would be affected by tick infection rates and dog behavior. In both Counties, dogs are not allowed to run free but must be kept close to their owner. Wild dogs would not be tested since someone must pay for the Vetrinary care.

The rates should roughly track. Since dogs are not familiar with ticks and Lyme disease, their infections should loosely track tick infection rates due to their inability to avoid bites and remove the ticks. Owners are often unable to detect tick bites early due to even short hair. But the estimates of tick infection rates with Lyme in New London are about 10 times that of Sonoma based on Vector Control and research studies in CT. Sonoma does have rural areas with tick infection rates as high as 41% according to research by Bob Lane. These areas would be close to New London tick infection rates but vary dramatically over small geographic distances. The higher rates do appear in areas where people visit with dogs such as parks. The dog behavior factor versus humans should not affect these rates per 100,000 but rather the absolute numbers. Dog and human behavior in Sonoma and New London should not differ by a large factor impacting the ratios. Only 9 cases of Human Lyme disease were reported in Sonoma CA in 2014.

http://www.sonoma-county.org/health/pub ... isease.pdf

New London CT had 2346 cases of Human Lyme disease in 2014.

http://www.ct.gov/dph/lib/dph/infectiou ... s_2014.pdf

So there was a 260:1 ratio of New London Lyme disease human cases in 2014 or 260 times that of Sonoma County. The ten times CDC reporting error should apply equally to both not affecting the population rate ratio. This means there were 7038 Lyme infections per 100,000 people reported in New London in 2014. There were 45 Lyme infections reported per 100,000 people in Sonoma in 2014! That is a discordance of 156 times not explained by either tick infection rates or dog risk behavior. This discordance in rates by population cannot be explained by anything except diagnosis, testing and reporting behavior by Sonoma doctors versus Veterinary doctors.

It seems rather odd that this discordance has not been noted and explored. Here is a list of some hypothesis that could explain this discordance between Sonoma and New London.

• Sonoma doctors are not educated properly and are less likely to entertain a Lyme diagnosis.

• Sonoma is highly biased by the Kaiser HMO which promotes minimizing Lyme and drives local doctor education

• The California Department of Health ( CDPH) also promotes minimizing Lyme

• The Sonoma County Medical Association promotes minimizing Lyme

• The Sonoma County Laboratory promotes minimizing Lyme

• The CDC 2T testing is based on the East Coast B31 antigens which are genetically distant from California strains and species lowering test sensitivity when utilized

• Sonoma Veterinary doctors do not seem caught in the minimization problem or controversies and only care about dog health

• Veterinary doctors use the C6 ELISA as in the SNAP 4DX Plus rather than the old 1994 B31 CDC 2T test

• Sonoma doctors use the IGenex lab more often which is disallowed from reporting in Sonoma so positive diagnosis are inored

One recent example of how Kaiser Santa Rosa, the CDPH and Sonoma County Laboratory publicize education to doctors that minimizes both Lyme and all tick borne diseases. The educational article here is full of errors all seemingly lowering the impression of risk of TBDs.

http://www.nbcms.org/about-us/sonoma-co ... &tabid=747

Sonoma has rural areas with tick infection rates often from 22-41% according to research by Bob Lane and Yvette Girard of UC Berkeley. They also found one rural community in Mendocino just 25 miles north of Sonoma had an unbelievable 23.7% human infection rate. That one study alone had over 25 people in one small community with a Lyme infection detected by UCB's optimized PCR testing. Human Lyme disease in Sonoma County and much of the California North Coast is simply being missed by poorly educated doctors, poor testing, poor education and bad judgement. It doesn't take a rocket scientist to see something is not right. But doctors are not always the brighest bulbs or most imaginative types on the big tree with some notable exceptions.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3067736/

Food for thought. Another Red Flag. Ther are certainly some reasonable factors but 156 times? I don't buy it and its shameful that its not been mentioned or investigated like the CDC 10X error or tick testing effort. I wonder why? Your guess?
The greater the ignorance, the greater the dogmatism.

Attributed to William Osler, 1902

duncan
Posts: 1370
Joined: Wed 5 Sep 2012 18:48

Re: Human versus Dog Infection rate Discordance

Post by duncan » Sat 3 Oct 2015 2:28

Lots of interesting data here, LHCTom.

I suspect most here are familiar with that billionaire in the UK whose entire family has been diagnosed with Lyme. I'm not really confident of the prevalence figures in the UK for Lyme in humans. I wonder if they looked at canine Bb incidence if they might get a more accurate indicator of the risks at play.

hv808ct
Posts: 256
Joined: Wed 30 Jul 2008 4:11

Re: Human versus Dog Infection rate Discordance

Post by hv808ct » Sat 3 Oct 2015 17:48

Human versus Dog Infection rate Discordance
Post by LHCTom » Fri 2 Oct 2015 19:17
There exists a suspicious disparity or discordance between veterinarian reported Lyme disease in dogs and that of humans between Sonoma County California and New London Connecticut. New London is near the worst Lyme rates in the US and near where Lyme was first discovered in the US in 1975. Since dogs are good surrogates for their owners, one would expect dog infection rates to roughly match human rates by population. The hypothesis is that the dog infection rate should remain similar by population as the human rate while absolute numbers should vary by factors like dog behavior and inability to avoid and have ticks removed due to hair.

Baker Institute at Cornell:

In contrast to human cases of Lyme disease, where three different stages are well known, Lyme disease in dogs is primarily and acute or subacute arthritis. The acute form may be transient and may recur in some cases. The devastating chronic stage in humans with systemic disease has rarely been seen in dogs.

The proportion of infected dogs that develop clinical disease is far smaller than it is for humans. Serological studies suggest that while more than 75 percent of the dog population in hyper-endemic areas may be exposed to infected ticks, only about five percent of those exposed actually develop clinical signs that might be attributable to Lyme disease. Within endemic areas, “hot spots” of tick infestation where dogs have a much greater probability of acquiring an infection are intermingled with non-infested areas where the habitat is not favorable to the vector tick. There may be age, breed, and genetic differences in the susceptibility of dogs to Lyme disease, but little is known yet about these factors.

Our studies indicate that bitches that become infected while pregnant do not transmit infection to their fetuses. Furthermore, our studies have shown no evidence that the pups of an infected bitch acquire the infection from her after birth.

…if a dog has never been in an area known to be infected with Ixodes ticks carrying B. burgdorferi, it is very unlikely that the dog will have Lyme disease. A diagnosis based on clinical signs often remains questionable, for there are several other conditions, such as immune-mediated disease and rheumatoid arthritis, that cause lameness and pain in dogs. A positive antibody titer alone tells very little. Many dogs with high antibody titers fail to have clinical signs. The presence of a specific type of antibody (IgM) is a good indicator of recent infection with some diseases, but in cases of Lyme disease, IgM antibodies persist.

duncan
Posts: 1370
Joined: Wed 5 Sep 2012 18:48

Re: Human versus Dog Infection rate Discordance

Post by duncan » Sat 3 Oct 2015 18:19

"The presence of a specific type of antibody (IgM) is a good indicator of recent infection with some diseases, but in cases of Lyme disease, IgM antibodies persist."

Someone, anyone, please inform the IDSA.

User avatar
LHCTom
Posts: 341
Joined: Mon 22 Oct 2012 4:18

Re: Human versus Dog Infection rate Discordance

Post by LHCTom » Sat 3 Oct 2015 19:39

HV808ct

Baker? :bonk:

Your hero? :shock:

Please focus and don't lose sight of the issue. Interesting commentary by Baker but it doesn't seem terribly relevant and an odd mixture of ideas and speculation. Does he talk with dogs? Maybe if I simplify the concept, step by step, you can use your brilliance to explain rather than quote something irrelevant and based on more guesswork rather than the scientific method.

In my experience living among infected ticks on rural acreage and as a dog owner throughout my life, VETs lean heavily on exposure probability and antibody based testing to diagnose an illness. The input from the owner is typically something minimal like "he is not behaving normally". They use tests like the Idexx SNAP 4Dx Plus which can detect Dirofilaria immitis, antibodies to Borrelia burgdorferi, Anaplasma phagocytophilum, Anaplasma platys, Ehrlichia canis and Ehrlichia ewingii in canine serum, plasma or whole blood.

The serological tests indicate antibodies to the pathogen are probably present based on the sensitivity and specificity. Bo the CDC 2T which is just an ELISA based on B31 based antigens followed by a Western Blot which is also based on B31 based antigens except electrophoresis allows the individual surface antigens to be separated, identified and quantified to some degree. The claimed ability to correctly detect antibodies to a specific pathogen typically is in the 90% range. That if correct and used uniformly should not cause a 2 orders of magnitude error.

The C6 is similar to the Lyme ELISA except it identifies a specific highly conserved peptide amino acid sequence found on the Vlse surface protein. Both the CDC 2T and C6 ELISA have been shown to have an ability to detect an infection once the adaptive immune system has had time to produce sufficient antibodies to detect. Most ID and Immunologist claim these antibodies remain detectable for years after the initiation of the infection. The CDC has debated whether to use the C6 rather than the 2T but one study indicated a less than 1% specificity. Somebody needs to explain to them arguing about 1% is silly when the accuracy of the research is most likely far worse than 1%. One of the earliest concept taught to engineering students is if you have a meter with 5% accuracy and hasn't been calibrated against a reference standard lately, 1% means nothing. Nothing. The reason is obvious.

Human and Dog immune systems use the same mechanisms to identify pathogens such that B cells are able to produce antibodies that bind to immunodominant epitopes on the pathogens surface antigens. This allows dogs to be tested for various pathogens through ELISA's that target antibodies to a particular pathogen. The same antibody techniques are used by Doctors and Veterinarians to see if antibodies are present suggesting either an active infection or past exposure.

Whether one looks at short term infection or exposure rates that led to antibodies or from a cumulative multi-year exposure point of view, the human and dog ratio's I explained should not be over 2 orders of magnitude in discordance unless something is very different in the diagnosis/testing of each.

A dog certainly can't complain of non-specific symptoms such as the found in PTLDS or Neuroborreliosis. Its silly to claim "devastating chronic stage in humans with systemic disease has rarely been seen in dogs.". Even when my dog seems listless and keeps collapsing his rear hind section due to some neurological disorder, his desire to please tends to mask whether he is ill or tired or ??. My dog doesn't need to work and function beyond pleasing the owner. He will go for a walk and try and jump and play but often collapses and appears bewildered and tries again after collapsing and dragging his back legs.

Its clear he has a neurological disorder that affects his back legs but its very difficult to know how he is feeling. When I ask him, he looks at me with a quizzical expression. He could be experiencing fatigue and a flu-like feeling and unless its severe, he will try and try. So a VET is able to get only minimal input about symptoms beyond "he doesn't seem like himself". So they rely on history of exposure to vectors like ticks and testing without much symptom input.
Our studies indicate that bitches that become infected while pregnant do not transmit infection to their fetuses. Furthermore, our studies have shown no evidence that the pups of an infected bitch acquire the infection from her after birth.
What does this have to with anything? Remember the 1% thing? Dogs and people born with Lyme represent an infinitesimal element of this problem. The meter has no infinitesimal accuracy.
…if a dog has never been in an area known to be infected with Ixodes ticks carrying B. burgdorferi, it is very unlikely that the dog will have Lyme disease
.

Duh. Brilliant observation Watson!
pain in dogs
Last time I asked my dog his level of pain he couldn't remember the number system so gave me that quizzical look.
Within endemic areas, “hot spots” of tick infestation where dogs have a much greater probability of acquiring an infection are intermingled with non-infested areas where the habitat is not favorable to the vector tick.
This applies to their human owners but this FACTOR should not change the ratio between Sonoma, CA and New London CT. This and most of this is only changes marginally between Sonoma and New London Counties. As I said, one study by the respected scientist Bob Lane found one rural hotspot 25 miles north of where I live in Sonoma had a 41% nymphal tick infection rate and 23% of the people who lived in this rural setting tested positive for Bb. Nobody tested their dogs but I'm sure they were comparable or worse. But this does not impact the ratios.
There may be age, breed, and genetic differences in the susceptibility of dogs to Lyme disease, but little is known yet about these factors.
Speculation? Sounds like Baker making an untested hypothesis and placing it in an argument as though its truth.

Ok. Make any sense? Probably not but others might get it. Its the ratios and they should not be off by 2 1/2 orders of magnitude without explanation of the underlying reasons. Just think of the effort made toward tick testing by Vector Control and researchers. Ok that indicates the pathogen is in the vectors n the area and gives a general idea of population probability. But the hotspot argument explains why this is invalid for any one person. If a doctor rules out Lyme because the area is not a highly endemic area and only an endemic area, it would be wise to ask about the areas they visit. One fourth of the people in this one rural area were infected at one time. That's quite a few. If a person lived in that area or visited it because a friend lived there, their risk goes way up. Sonoma doctors rarely ask questions about exposure.

If you cannot understand why the dogs versus human ratios indicate a problem related to exposure probability beliefs and assumptions in doctors, you need to take a course in "Common Sense 101" and "Probability and Statistics" at the same time while thinking about how they might apply. This assumes you even care about truth. If you are so wed to your beliefs you have lost sight of truth, maybe this will help someone else. Actually, that's why I respond. The probability you will EVER be open to truth is close to zero. Very very close and immeasurable. Sorta like trying to pin down the velocity and location of a subatomic particle at the same time.

Thanks for giving me a reason for throwing out some interesting ideas that might help someone somewhere.
The greater the ignorance, the greater the dogmatism.

Attributed to William Osler, 1902

hv808ct
Posts: 256
Joined: Wed 30 Jul 2008 4:11

Re: Human versus Dog Infection rate Discordance

Post by hv808ct » Sun 4 Oct 2015 1:23

Re: Human versus Dog Infection rate Discordance
Post by LHCTom » Sat 3 Oct 2015 19:39

Baker?
Your hero?

Please focus and don't lose sight of the issue. Interesting commentary by Baker but it doesn't seem terribly relevant and an odd mixture of ideas and speculation. Does he talk with dogs?
You know Baker is a place, not a person? It’s part of the Cornell Vet Sch and has been funded by NIH for many years now. It’s the go-to place for questions about LD and canines.

duncan
Posts: 1370
Joined: Wed 5 Sep 2012 18:48

Re: Human versus Dog Infection rate Discordance

Post by duncan » Sun 4 Oct 2015 14:42

Well, if it's funded by the NIH, and it has anything to do with Lyme, then I suspect most of us here understand its value.

User avatar
ChronicLyme19
Posts: 564
Joined: Mon 11 Aug 2014 17:42
Location: NY, USA

Re: Human versus Dog Infection rate Discordance

Post by ChronicLyme19 » Sun 4 Oct 2015 23:10

LHCTom wrote:The serological tests indicate antibodies to the pathogen are probably present based on the sensitivity and specificity. Bo the CDC 2T which is just an ELISA based on B31 based antigens followed by a Western Blot which is also based on B31 based antigens except electrophoresis allows the individual surface antigens to be separated, identified and quantified to some degree. The claimed ability to correctly detect antibodies to a specific pathogen typically is in the 90% range. That if correct and used uniformly should not cause a 2 orders of magnitude error.
A very nice analysis LHCTom. I agree it is probably very unlikely that behavioral changes between the two populations would lead to orders of magnitude difference in the rates, absolute numbers sure, but the rates definitely less likely. I think it is very plausible that the rate of human infection to dog infection would stay relatively constant on both costs, and most likely in CA is several under reported.

For what it's worth, it's also severely under-reported on the east coast endemic areas as well. Take my town, which is about 16K people, in endemic Dutchess County, NY. For all of 2014, only FIVE cases of Lyme were reported by the county DOH. FIVE. We've done some math before and we think the rates should be more like 5-10%, which would mean on the conservative side, 800 people in our town per year. 5 is also orders of magnitude off from the 800 or so we expect. Factors we've identified influencing the poor reporting include, poor testing, borrelia species/strain (Westchester just south of us has miyamotoi rates close to 7%), lack of testing/reporting fatigue, lack of funding by the state/county DOH, and fear by the doctors to report.
LHCTom wrote:The C6 is similar to the Lyme ELISA except it identifies a specific highly conserved peptide amino acid sequence found on the Vlse surface protein. Both the CDC 2T and C6 ELISA have been shown to have an ability to detect an infection once the adaptive immune system has had time to produce sufficient antibodies to detect. Most ID and Immunologist claim these antibodies remain detectable for years after the initiation of the infection. The CDC has debated whether to use the C6 rather than the 2T but one study indicated a less than 1% specificity. Somebody needs to explain to them arguing about 1% is silly when the accuracy of the research is most likely far worse than 1%. One of the earliest concept taught to engineering students is if you have a meter with 5% accuracy and hasn't been calibrated against a reference standard lately, 1% means nothing. Nothing. The reason is obvious.
We've heard word that some folks in the county/state DOH are pushing to get the C6 included for reporting statistics. I hope that is true.
LHCTom wrote: Whether one looks at short term infection or exposure rates that led to antibodies or from a cumulative multi-year exposure point of view, the human and dog ratio's I explained should not be over 2 orders of magnitude in discordance unless something is very different in the diagnosis/testing of each.
Agreed.
Half of what you are taught is incorrect, but which half? What if there's another half missing?

Post Reply