Today's culprit is an article from last year in the Huffington Post last year. Its title is:

Lyme Disease: The Perfect Storm is Heading Our Way

The author is Leo Galland, M.D. You can read it for yourself at http://www.huffingtonpost.com/leo-galland-md/lyme-disease_b_1429984.html

The reason I'm discussing this article is because it made a prediction that did not come true. The author and other scientists predicted an increase in the number of lyme disease cases based on two factors:

  1. 1.) a warm 2011/2012 Winter, and
  2. 2.) a large acorn crop in 2010 followed by a smaller-than-usual acorn crop in 2011.

Ticks live for two years, assuming they survive their larval feeding and their one and only Winter attached to a host. Galland reasoned that more ticks would survive in

an unusually warm winter, which left deer ticks alive, hungry and looking for a meal.

His statement about acorns was the following:

The mice feed on acorns and store them for winter. The fall of 2010 brought a bumper crop of acorns, which led to a surge in the mouse population and created abundant homes for tick larvae last spring (2011). In the fall of 2011 the acorn crop was the smallest it's been in two decades, decimating the mouse population over the winter and leaving a huge number of displaced nymphs that are looking for warm-blooded hosts, like humans.

To be frank, Galland presented his prediction as sensational news and I confess that it annoyed me. He began his article with:

Blood-sucking ticks coming to a field and forest near you. That may sound like the latest horror film, but unfortunately it is a reality due to a surge in ticks that spread Lyme disease this spring.

To make matters worse, he went on to brag:

media interest in Lyme disease appears to be growing with the threat. At the start of the month I was interviewed on Martha Stewart Living Radio about Lyme disease.

Galland did include a long list of references, three of which looked at the ecology of Lyme disease vectors. One of these, Schauber et al. (2005), stated in its abstract:

Acorn production and mouse abundance....were the strongest predictors (r =0.78)....(of) Lyme disease incidence

The above makes a case for correlations between acorns, mice and ticks and a regression coefficient of 0.78 is not bad. What's missing is the formal quantification of the two year lag for the acorn crop in this paper or in the other two ecology references. Where then did Galland get the details on the acorn-mice-tick-Lyme disease chain? It's possible that he used Ostfeld et al. (2006), which is a highly cited article. He might have gotten it out of Ostfeld's book on the ecology of Lyme disease, a book I don't have access to and can't afford to buy. He might even have gotten it from a blog interview of Ostfeld last year on a website called pilladvised.com. Overall, Ostfeld is the go-to source for info on the acorn-mice-tick-Lyme disease connections. Let's look at Ostfeld et al. (2006).

Ostfeld et al. (2006) is online at PubMed (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1457019/, accessed May 6, 2013), Ostfeld et al. (2006) looked at multiple weather effects as well as deer populations, chipmunk populations, mice populations, larval tick populations, nymph tick populations, infection density of tick nymphs, and acorn abundance. Ostfeld et al. (2006) did mention some interesting correlations in their paper from previous studies.. One interesting one was:

In the laboratory, ticks experience high mortality when exposed to low humidity and high temperatures. Consequently, hot and/or dry springs and summers have been postulated to reduce subsequent nymphal tick densities and Lyme-disease risk.

Another correlation mentioned was:

Because adult I. scapularis (deer ticks) feed predominantly on white-tailed deer, much research has evaluated the impact of variation in abundance of deer on abundance of ticks. When deer are eliminated from some habitats by hunting or fencing, the abundance of ticks typically is strongly reduced. Studies comparing natural variation in deer abundance with that in tick abundance are less conclusive; some have shown strong associations, whereas others have not.

The Ostfeld et al. (2006) study was not a single variable, laboratory-only effort. The conclusions of Ostfeld et al. (2006) were based on the statistical analysis and multivarient modeling of 13 years with of data collection in oak-containing forests in upstate New York. Their work was very data-driven and they used their 13 years of data to tightly constrain their models. When all was said and done, they noted that:

Among all univariate models, NIP (nymph infection prevalence, i.e. the numbers of Lyme carrying-juvenile ticks) responded only to the density of acorns in year t−2, and this relationship explained only 16% of the variance in NIP. Models incorporating the climate variables, deer variables, DOL (prior year's density of tick larva), mice, chipmunks, and total rodents performed no better than the means model. Because none of the independent variables other than acorns t −2 produced an improvement over the means model, there was no justification for testing multiple regression models.

In plain English, Ostfeld et al. (2006) found that Lyme disease risk from ticks correlated positively with an abundant acorn crop two years prior, but the correlation wasn't all that strong. They also noted that no other variables in their models caused statistically significant increases or decreases in Lyme disease risk. This was a surprise as far as deer were concerned because for years, people tried to correlate Lyme disease with deer ticks and deer abundance.

Their concluding paragraph in the discussion of their results stated:

Climate, deer, and acorns each have been proposed as primary determinants of temporal variation in risk of human exposure to Lyme disease, as measured by abundance and Borrelia-infection prevalence in nymphal Ixodes ticks. Using a model comparison approach and a 13-year dataset, we found weak support for climate variables, no support for deer, and strong support for an effect of acorns, mediated by acorn effects on white-footed mice and eastern chipmunks, which host many larval ticks and are competent reservoirs for B. burgdorferi.

It is a safe assumption that Galland had one of Ostfeld's papers or books on hand when he wrote about the acorn to mice to tick connections.

Blog note: Borrelia burgdorferi is ...

The other ecological study cited by Galland was Keesing et al. (2009) (http://rspb.royalsocietypublishing.org/content/early/2009/08/19/rspb.2009.1159.full, accessed May 6, 2013). In this study, the researchers caught white-footed mice, eastern chipmunks, grey squirrels, opossums, veeries and catbirds. They then penned their animals and deliberately introduced larval ticks to the captive population. When the ticks fell off after 3 days (the average duration of feeding), the researchers would remove one species and reintroduce more larval ticks to the remaining animals. This experiment was designed to see how that species removal would affect the redistribution and survival of ticks among the other five. It was a large and complicated experiment and I admit here than I glossed over some of the details for the sake of brevity.

Here are some Keesing et al. (2009) results: the common white-footed field mouse was a tick host par excellence. Fifty percent of the ticks that tried to attach on mice stayed on mice until they were engorged and fell off. Other animals were more aggressive with grooming, biting or rubbing ticks off. Opossums and squirrels were especially good at removing ticks, managing to get rid of 96% to 83% of all ticks, respectively, that tried to attach and feed. Most of ticks that were groomed off were consumed by their hosts. For all the other animals in the study, between 70% and 80% of the ticks failed to attach.

Surviving ticks

This figure is from Keesing et al. (2009).

Keesing et al. took their data on tick attachment and survival and used it to model the risk of Lyme disease. The model results were somewhat surprising. Removing mice caused the density of infected nymphs (DIN) to fall. Removing opossums caused the DIN to increase. With the caveat that the experimental set-up was a closed system, the Keesing et al. results for mice appear to be the exact opposite of what Galland stated in his article.

Predictions, scientific or otherwise, can only be validated by data after the fact. Galland's prediction of a "perfect storm" of Lyme disease for 2012 should be compared to the actual numbers of reported cases. If Galland was right, there should be an observable increase of Lyme disease in 2012. We have this data but first we need to do something about the change in counting Lyme disease cases that occurred in 2008.

Prior to 2008, Lyme disease statistics were reported only for confirmed cases. Starting in 2008, both confirmed plus probable cases were reported. These numbers are available for the years 2008 thru 2011. Validated statistics for 2012 are not available but an unvalidated confirmed+probable number was reported in the CDC's Morbidity and Mortality Weekly report for the last week in 2012.

Using the data from 2008 thru 2011, we can calculate an average ratio of confirmed+probable cases to confirmed only cases. The average ratio is 1.28. We can use this ratio to estimate confirmed+probable cases prior to 2008 and also to estimate a confirmed only number for 2012. All these numbers are included in the table below.

tickstatscalc.png

One can see clearly from the table that Galland's and Ostfeld's predictions of Lyme disease disaster were wrong. Lyme disease occurrences actually dropped in 2012 with respect to the year before. So what went wrong with the prediction? Nobody knows! The problem here is that we're dealing with a truly complex system where some hitherto unidentified variable just overrode the acorn-mice-tick correlation. This tells us that there are still some unconstrained variables out there waiting to be discovered. The cause could be anything that's plausible. As just one example, there could have been a tick die-off because of the extremely dry and hot weather during the Summer of 2012, the third hottest Summer ever recorded. Or perhaps there really was a mouse die-off from inadequate forage during the Winter of 2011/2012, as predicted, leading ticks to search for new hosts. But as Keesing et al. (2009) suggested, those ticks could have attempted attachment with tick-killing animals like squirrels and opossums, leading to the death of those ticks and thus reducing tick numbers and Lyme disease risks. One can make many plausible hypotheses to explain the unexpected decline in Lyme disease cases in 2012, but without more data on possible new variables, it's nothing but speculation for now.

For an MD, Galland's no ecologist. His article never mentioned the west coast blacklegged tick, which is the Lyme disease carrier west of the Rockies. Galland also showed no awareness that some regions in the USA are free of Lyme disease or that Lyme disease hot spots include the northeast, Minnesota, and Wisconsin. These are sins of omission which become apparent when we look at and compare a blog interview of Ostfeld posted on the pilladvised.com website (ibid.). It looks like Galland borrowed heavily from the Ostfeld's blog interview (ibid.), including some section titles and references verbatim. Ostfeld qualified his tick surge estimate as valid only for the northeast and only for the deer tick, something Gallard did not do. Missing these caveats was not great journalism on the part of Galland, in my not so humble opinion.

Screen-shot-2012-02-01-at-11.05.30-AM.png

I have to wonder if Galland and Ostfeld are in cahoots with each other since the pilladvised.com website is owned by Galland and Ostfeld posts there..

I have a personal feeling that Galland was after sensationalism deliberately. Here's why I think so:

  1. His title and opening paragraphs used sensational wording, as we noted earlier (sensational claims)
  2. He cherry picked research results to support his hyped-up statements (data massaging)
  3. His reference section was blatantly cut and paste (false impression, bad style)
  4. He made an alarmist statement regarding the alleged failure of Lyme disease treatments (sensational claims)
  5. He opined that chronic Lyme disease is uncurable (sensational claim)
  6. He posted his article to a news aggregater with a reputation for pseudo-science and mostly amateur efforts with no editorial review.

Some of the practices in the above list speak for themselves but other need some explaining. We need to look closely at Galland's statement that the standard Lyme disease treatment of antibiotics is a failure. His evidence consists of journal articles reporting on failed treatments. He lists thirteen such articles in his references section. He ignores studies that contradict his stance on treatment failure: this is another sin of omission, one that is significant. While he told no outright lies, Galland's presentation of his failure data and his lack of discussion on alternative hypothesis made it appear that Lyme disease was an unstoppable epidemic with poor chances of successful treatment.

To test Galland's assertions, I visited the Center for Disease Control (CDC) website looking for data. The CDC is the best place to obtain data on all sorts of things like the incidence of various diseases and the causes of death, to name two examples. It's a gold mine of robust data. The CDC reports that 10% to 20% of people with Lyme disease do not respond to treatment by antibiotics in the short term (http://www.cdc.gov/lyme/treatment/index.html, accessed May 6, 2013). Antibiotics administered soon after infection are usually successful, whereas antibiotics administered many weeks to months after infection have a greater risk of failing http://www.cdc.gov/lyme/treatment/prolonged/index.html, accessed May 8, 2013). In a soundbite: antibiotics administered soon after infection have an 80% probability for successful recovery. That's not a failure. Antibiotics work for most people in recovering from the early stages of Lyme disease.

I opine that Galland cherry-picked those 13 journal articles because they supported his allegation of treatment failure. It's a form of massaging the data to fit your hypothesis and it's bad science.

As to whether chronic Lyme disease is uncurable, the CDC notes that most people with persistent symptoms eventually get better on their own regardless of whether antibiotics are administered; however, the recovery rate can be extremely slow for some people (http://www.cdc.gov/lyme/postLDS/index.html, accessed May 7, 2013). The bottom line is that Lyme disease, like shingles and syphilis, is an infectious disease which should be treated as soon as possible after infection. Like late-stage syphilis, chronic Lyme disease may be difficult to impossible to treat. For most people, it looks like the only effective cure for chronic Lyme disease is time; but there are still a few sufferers left over who never improve.

Stuff like Galland's article does real damage. When a journalist or blogger writes up a scientific result as the latest and greatest breakthrough ever, and then caps it with a sensational title, readers become rather cynical about science reporting and about science in general. Every time the content of an article doesn't live up to its title, another chip of respect for science is lost. But it's not scientists who are doing this - it's journalists. Slapping an outrageous reader-bait title on a science-based article has become the new norm for internet reporting. It happens all the time. There are so many misleading and sensational headlines on the internet that they become meaningless rather quickly. It's like a real world varient of the boy who cried wolf. When someone makes a science-based prediction, like last year's tick prediction, and it fails, the worth and credibility of science is further eroded. The only thing sensationalism really achieves in the news media is the attraction of readers toward webpages with advertising or phishing scams. The content and the presentation of reporting has been bent to maximize viewer ratings and advertising exposure. There's really very little real news anymore; it's all thinly disguised entertainment and the facts be damned.

Why am I suddenly reminded of the razzle-dazzle song from Chicago?