One has to be able to distinguish between make-believe and science

Round 3: University of Toronto plays dumb about #FismansFraud

When did incompetence become a superpower in our institutions?

Trying to get the University of Toronto to do the right thing and investigate #FismansFraud is like pulling teeth. Their resistance is staggering. Their latest tactic: to play stupid, deflect and hope I go away.

Am I to believe that the office responsible for oversight in research integrity doesn’t understand the definition of fraud and cannot recognize a textbook case of it?

No.

Quite frankly, I think it’s more of a gaslighting technique than incompetence. But we’ve witnessed so much of both from our “experts” and authority figures during this pandemic that we can’t rule out either.

So the dance continues.

Their latest letter of inaction has forced me to step things up yet another notch, a pattern they should be getting use to. I’ve included my response below for those interested. I had some fun with this one and I daresay they won’t enjoy the read as much as I enjoyed writing it.

Why am I going through such great effort to get this one study investigated and retracted when there are so many other egregious violations happening within our scientific community?

First, because data fabrication and falsification are two of the most serious acts of scientific misconduct and the Fisman et al. study is a blatant example of it. Under no circumstances should such a faux study be supported in a legitimate scientific institution. If it stands, it indicates their scientific integrity process is broken.

Second, the fraudulent findings were quickly and widely disseminated by the media in order to create further divide and scapegoat an innocent segment of society. Worse, our government attempted to use the study to justify human rights violations and further restrictions.

Third, we’re likely to see more of these types of modelling-based shams in the days, weeks and months to come.

Our government has indicated that it’s not willing to give up its pandemic charade. The Liberals have stated outright that they will not hesitate to bring back the mandates and restrictions. Moreover, injecting 100% of the population with the rushed-to-market vaccines has been their singular goal throughout this mess. Not to mention that they’ve purchased many years worth of the vaccines. They are too heavily invested to let this drop. It does us no good to pretend they’re not using every waking hour to figure out a way to get the public to buy into more injections this coming fall with required top-up boosters in perpetuity.

The problem is, objective science and real data doesn’t support their agenda. Not at all. As such, they’ll have to rely on make-believe under the guise of science and one of the easiest ways to do that is through contrived mathematical modelling. This approach has been proven effective throughout the pandemic, as witnessed by Ferguson’s flawed model that triggered the global lockdowns and the ridiculously inaccurate modelling put forth by our very own Ontario Covid-19 Science Advisory Table on which Fisman served.

I suspect the government will be in need of additional modelling services come fall. Perhaps Fisman is on standby?

We must address this now.

Here’s my response to U of T’s lame letter. Do you think the allegations have been made simple enough for them to understand? Let me know what you think in the comments section.

Round 3: Letter to U of T — this is what fraud looks like

Re: Research Misconduct Submission: David N. Fisman, Ashleigh R. Tuite & Afia Amoako

Dear Professor Lori Ferris and the Research and Oversight Committee:

Thank you for your letter on June 21st which sheds light on your interpretation of the allegations of scientific misconduct against David N. Fisman, Ashleigh R. Tuite and Afia Amoako. I have considered carefully the arguments presented therein and have found that while they provide support for the use of mathematical modelling in science – which is not in question – they do not address, nor provide defense against, the allegations of fraud at hand. I have addressed each of your comments below to further clarify the acts of misconduct at issue in hopes we can reach a satisfactory resolution.

The use of models to simulate hypothetical outcomes and explore real-world phenomena can, when properly constructed, provide great insight into real-world issues. Such simulations do not, in and of themselves, constitute data fabrication. That is not in dispute. However, passing off simulated data as a true reflection of events without any real-world validation does constitute data fabrication by its very definition. The violation is especially egregious when real-world observations are readily available and such data completely contradicts the contrived results, as in the present case.

Validation of results is a key component of mathematical modelling. It provides a reality-check to ensure the integrity of the findings when making inferences and predictions about real-world scenarios. Moreover, model validation helps eliminate fraud by preventing the complete fabrication of data. That is, validation prevents the kind of fraud we witnessed in the Fisman et al. study where a model is contrived merely to satisfy the highly biased, preconceived notions of the authors.

As you mentioned in your letter, the Fisman et al. paper clearly states that the study is based on a model. However, the fact that a model simulation was the basis for the study in no way negates the need to validate the model and the results before making inferences about real-world outcomes. Indeed, well-established methodology calls for such validation. Certainly, the authors know this since two of them, namely David Fisman and Ashleigh Tuite, specialize in mathematical modelling. Moreover, the main author cites mathematical epidemiology as a primary teaching responsibility.

Yet, no real-world data was modelled. There was no benchmarking or validation of any kind. Hence, no scientifically valid inferences about real-world events are possible from their study. Regardless, the authors go on to state their fabricated results as fact:

“…we found that the choices made by people who forgo vaccination contribute disproportionately to risk among those who do get vaccinated.”

“The fact that this excess contribution to risk cannot be mitigated by high like-with-like mixing undermines the assertion that vaccine choice is best left to the individual and supports strong public actions aimed at enhancing vaccine uptake and limiting access to public spaces for unvaccinated people, because risk cannot be considered “self-regarding.” There is ample precedent for public health regulation that protects the wider community from acquisition of communicable diseases, even if this protection comes at a cost of individual freedom.”

The above quotes, taken directly from the Fisman et al. paper, assert a finding regarding choices made by people despite a complete absence of people or real-world observations in their study. They go on to state their findings as fact, falsely claim that their fictitious fact supports “strong public actions” and they even demonstrate awareness that imposing such actions based on their fiction comes at a cost to individual freedom.

Note that the model was not used to project the future, baseline conditions were set to simulate the past. The authors simulated an epidemic that had already occurred this past winter, got it wrong, then proceeded to overwrite history by stating that they “found that the choices made by people who forgo vaccination contribute disproportionately to risk” when, in actuality, incident rates were found to be disproportionately higher amongst vaccinated individuals.     

To date, the authors have failed to clarify that their “findings” do not relate to real events and are, in fact, opposite reality. This despite many researchers, scientists and others calling on the authors to do so. Instead, the authors have staunchly stood behind their contrived results to the detriment of society. Concerns provided to CMAJ from fellow researchers have been met with hand-waving dismissive responses, deflections and further falsehoods.

Knowingly stating fiction as fact is fraudulent. Whatever caveats may have been included in the study do not change that. Once brought to the attention of the authors, the journal, and the university, corrections should have been made. Instead, all parties are finding reasons to turn their backs on their duty to uphold scientific integrity thus allowing these false inferences to propagate and cause further harm.

Your letter further justifies inaction by pointing out that the Fisman et al. paper clearly states the limitations of the model. This is not the case as demonstrated by the many issues raised by myself along with other scientists and researchers following the publication. The only weakness mentioned in the paper is that the model “does not precisely simulate a real-world pandemic process in all its complexity” – a statement true of any mathematical model. In true form, the authors go on to provide a single, one-sided example of this weakness, namely that not all the benefits of vaccination were captured. They use this “weakness” to falsely assert that they likely underestimated vaccine benefits, once again without reference to real-world observations that show the opposite.

Far from a declaration of the model’s inadequacies, the authors’ statement of limitation serves to propel and even magnify the fraudulent results of the study and bolster a political agenda.

You state in your letter that the authors’ model is publicly available for any interested parties to review, which I and other researchers have done. Along with my initial letter, I provided a 20-page review of the study that included a number of critical limitations with the contrived model, none of which are mentioned in the Fisman et al. paper. I encourage you to read my review if you haven’t done so already; it provides a compelling case of scientific misconduct. I have included it again with this submission, for your convenience.

While I have a strong background in mathematical modelling and can easily recognize the complete invalidity of the research, the erroneous findings were quickly taken up by media and widely distributed to the general public. Is it your view that the onus is on the public to validate findings published by University of Toronto professors? Is it wrong for the average layman to assume that a peer-reviewed study, written by tenured or seasoned professors and published in a Canadian medical journal, had gone through proper scientific rigor – including validation? Certainly, the process isn’t perfect and papers slip by that shouldn’t. However,

there is an assumption that when egregiously false and harmful findings are brought to the attention of the journal and the research institution, they will be corrected.

As you state, “models have value in science even if they cannot mirror the real world with 100% accuracy.” This is certainly true. However, when they fail to adhere to proper scientific methodology they can also be used to mislead. The gross transgressions that allowed for the fraud in the Fisman et al. study are not par for the course. It is not a matter of the model falling short of 100% accuracy. The model they present is both flawed and fraudulent to the point of causing significant societal harm.

One has to be able to distinguish between make-believe and science and there are many tools available to do that. If well-established methodology had been followed, as claimed in your letter, then this fraud would not have survived the process.

I have submitted criticisms of the study to CMAJ. Thus far, only submissions received within one week of the publication date have been posted to their website. I have not been provided information as to if or when the remaining submissions will be posted.

Given the serious nature of these concerns, I notified University of Toronto’s Research & Oversight Compliance Office of the allegations and I provided supporting documentation to back up my claims. After receiving the university’s initial response to my complaint, I apprised CIHR of the situation. They, along with CMAJ, have been cc’d in this correspondence.

To date, none of my concerns have been sufficiently addressed by the authors nor any of the three institutions involved – CMAJ, University of Toronto, CIHR. I hope this letter provides further clarification into the allegations of misconduct so that the university can thoroughly address all questions presented here and in previous correspondences and work to resolve all concerns as set forth in Sections 6, 7 & 8 of the University’s Framework to Address Allegations of Research Misconduct.

As articulated throughout this letter and detailed in previously submitted material, the issues at hand go far beyond debates of scientific merit, they concern fundamental research integrity.

The authors conducted a textbook example of data fabrication and falsification. There is no question their statements of fact are fraudulent with potential for great harm. If this is not worthy of proper investigation, one is left to question what is.

Mathematical modelling does not grant one a license to mislead, to ignore real-world data, to fabricate or falsify results and call them reality. To foster such deception under the guise of science does a disservice to the scientific field, the research institute and to society.

Once again, if you have any questions or would like further clarification of the allegations, I can be reached directly at xxx-xxx-xxxx.

I look forward to a thorough response to my many questions and concerns and to a satisfactory resolution.

Sincerely,

Regina Watteel                                                                               
BSc. Hons Math & Physics, MSc. PhD. Statistics