Monday 27 May 2013

Day 148: Raymond Duch and public polling

As the Montreal tobacco trials entered the third week of the "defence proof," a third expert witness was called in to speak to what is seen as the industry's main line of defence - smokers' awareness of the dangers of using their products.

Over the past two weeks, the trial has been presented with a "deluge" of old press clippings and other reports of information presented to Quebecers during the second half of the 20th century. These came in the form of reports by two historians, Jacques Lacoursière (for Rothmans, Benson and Hedges) and David H. Flaherty (for Imperial Tobacco) as well as the literally thousands of published reports that were the basis of their analyses.

Raymond Duch
Today's witness, Raymond Duch, is not an historian, but is instead a numbers man. As a quantitative political scientist, he focuses most of his research efforts on economic voting. (As he explained it to the court this morning, it would appear that our voting decisions are not driven by altruism or political philosophy. Sigh.)

Although he originally hails from The Peg, Mr. Duch's current home address is in Oxford, England where he is a fellow at Neuffield College. From his careful and sometimes hesitant answers, my guess is that he is not often called on to explain math at a court-room or non-Oxford level. He did well at what must be a primary objective: he kept Justice Riordan attentive and engaged.

Mr. Duch's mandate from JTI-Macdonald was to focus not on what information Quebecers were exposed to through the media, as the historians had done, but rather to review polling results to assess what the public remembered hearing and what they believed as a result.

His Report on Public Attitudes Associated with Smoking: Quebec and Canada was filed today under Exhibit 40062. (The source material will be filed under Exhibit 40064).

His is another contribution to the hefty research library assembled courtesy of these first tobacco industry witnesses. Now to find a way to make it available!

The 'Read-in' Rule

The practice adopted in this trial has been to accept the reports of expert witnesses "as read," which is to say that they do not need to be formally presented. The implementation of this practice has varied considerably, but in today's instance it meant that Mr. Duch presented the highlights of his findings to Justice Riordan in response to the questions fed to him by one of JTI-Macdonald's counsel, Doug Mitchell.

(It has been some weeks since we have seen Doug Mitchell in this court. With his convivial questions and occasional bon mot today, he seemed far more at ease today than earlier in the trial when he was handed the job of trying to knock the stuffing out of the plaintiff's witnesses and their testimony.)

For most of the day, the lawyer and witness simply followed the outline of the report. This allowed Mr. Duch to put on the record his views that:

* there were "remarkably high" levels of awareness of reports that smoking can cause cancer among Quebecers - even as early as 1954. "88% for Canada and 82% for Quebec - those are essentially everyone in the population. You don’t often get numbers that high." (Beyond 85%, he said, means "you really have the whole population".)

* although in early years Quebecers had lower levels of awareness than other Canadians, those differences disappeared by the mid 1980s.

* reports linking smoking and lung cancer were treated seriously, in part because of people's fears of cancer and also because the information came from highly credible sources. "The message will be listened to and will resonate – it will catch the attention of listeners."

* by the mid 1980s, Quebecers and Canadians were telling pollsters that they believed the link between smoking and lung cancer and other health problems.

* the trend in public acceptance of the health consequences showed a steady and consistent growth until the late 1990s when it neared 100%

* around 85% of Canadians reported their belief tobacco use was habit forming or hard to quit in the early 1980s, and that a decade later they agreed that it was "addictive."
Mr. Duch observed a steady increase over time in
agreement that smoking is a cause of lung cancer

Inside jobs and Outside jobs

By now the failure of industry witnesses to examine internal documents is a well-established pattern. We did not have to wait for the cross-examination to have the issue put to Mr. Duch, who is the third witness in a row to overlook the rich vein of research material that has now become part of the court record.

The examination had barely begun before Justice Riordan asked the witness why he had limited his inquiry to public polls.

Like those before him, Mr. Duch said the decision had been his to make. "I made an early decision to only include what I call publicly available data or survey data referenced in peer reviewed journals. I decided not to include survey data gathered by interested parties." 

"You don’t give credence to what was done by the companies? ... were [the polling companies] reputable firms?" Justice Riordan asked.

The companies were perfectly reputable, Mr. Duch said, but he "felt would undermine the credibility of the report" if he used their results. "Maybe it was a wrong call on my part, but I felt that it was the right call." 

The best defence is a good offence

Mr. Duch's report is the second expert survey report to be filed with the court. The first was by the plaintiff's witness, Christian Bourque, who testified in January and March.

The sources used by the two men for their reports are mutually exclusive. In yet another irony,  Mr. Duch, who works for the industry, looked only at public polling. Mr. Bourque, who is a public pollster (with Leger Marketing), looked only at the industry's private data.

Nor are their findings compatible. Mr. Duch found an "exceptionally high level" of awareness and indications of perceptions of risk.  Mr. Bourque found that "a good number" did not perceive smoking to be dangerous, and that even by 1990, one smoker in 5 did not clearly understand that smoking was dangerous for everyone.

It's not possible to square these two perspectives - and this afternoon Mr. Duch was invited to square off against the plaintiff's expert who was sitting at the back of the room listening, just as he had sat to observe Mr. Bourque's testimony.

"I don't want to personalize this," Mr. Duch began before launching into a review that was in many ways the mirror of the criticisms levelled against his work and that of other industry witnesses.

Mr. Bourque erred, he said, because he only looked at the industry records and was "innattentive to the large body of existing public opinion research" that "should have been addressed or referenced in the report." As a result, "his conclusion is an order of magnitude different to what the publicly available data suggests."

He criticized the questions that were used by Canadian Facts during its much-repeated survey, saying that they were too ambiguous, or "conflated" issues by combining ideas. "The data that comes from the survey is entirely unreliable." The presentation of results was also flawed, he said, as Mr. Bourque had improperly interpreted answers to what he saw as a factual question about smokers' views on how many cigarettes could be smoked before incurring health risks.

The whole exercise was so flawed, in his view, that Mr. Duch charged the plaintiff's witness with "misleading the court" and taking an "indefensible position."  Ouch!

The plaintiffs continued to maintain a calm silence throughout the day -- it was Justice Riordan who sounded like he was defending the plaintiff's witness when he pointed out that Mr. Bourque's mandate had been limited to the industry's internal polling.

Pre-emptive strikes

The first two historians to testify for the industry had suggested that the tobacco companies had little credibility, and that there denials of health risk would therefore have had little impact. Although Mr. Duch did not review any public polling data on this, he was asked to comment on an internal poll conducted by Imperial Tobacco that at first blush - and without the explanatory tables - might suggest that half of Canadians thought the industry was credible. (See "52%" figure circled in red from Exhibit 987.21).

Perhaps fearful that this information would be misunderstood if presented by the plaintiffs on cross-examination, Doug Mitchell invited Mr. Duch to explain that the question had only been asked of those who said "yes" to the question of whether the industry could make a case to refute health concerns.

The arithmetic was presented (52% x 31% = 16%) showing that only 16% of Canadians thought the industry was credible (about 1 in 6 people).

Mr. Duch was not asked to comment on the fact that among smokers (circled in yellow), one-quarter of smokers found the industry credible. (61% x 41% = 25%).

By today's 1.5 million Quebec smokers, that would be 375,000 individuals, almost the population of Quebec's third largest city, Laval.

A short week

With virtually no interruptions from the plaintiffs, and a witness who seemed to explain his work concisely and clearly, Mr. Mitchell had almost finished by the end of the day. The cross-examination will likely start early tomorrow and be finished by the end of the day.

If so, a short week will have become even shorter -- as the tobacco companies have still not provided any back up witnesses to fill the holes in the schedule should Wednesday now be an empty day.

Which brings us to the next installment in the saga of Justice Riordan's case management decision.

Yes, abusive. But process, not people.

An enhanced politesse has entered the discussions between Justice Riordan and the defence counsel following their objections  to his decision to limit their proof to 175 days (their schedule, which has not been made public, identified 300 more trial days).

The lawyers' speeches were prepared in advanced, and read in the most moderate tones. And so was Justice Riordan's response when it was provided at the beginning of the day.

The judge said he had reflected on the complaints, and clarified that his use of the word abusive "is in relation to excessive and unreasonable" schedule that was proposed. He said he had made no allusions to improper behaviour by individual lawyers.

He acknowledged that is ruling was an "unusual step." While he did not step down from his decision to impose a time limit, he said he would "welcome" the offer of modifications that the companies had made "should they materialize."

He specified that his concern with respect to class members was that the Imperial Tobacco would ask for individual medical histories and records, which he has previously refused. Appeals of future decisions to deny access to these records "could add months to the trial." He said it was not appropriate for him to put any objections under reserve (i.e. force the class members to divulge their histories and then strike them from the record later).

He suggested either holding one week of hearings on class members (to get objections on the record) or having a test case identified earlier. "This would allow the appeal case to proceed."

He responded to the comments about his impartiality by stating that there was no basis to see his ruling as suggesting he had made any decisions on the case. "Pieces of the puzzles are still being laid out on the table and I do not intend to start assembling them until the end."

Justice Riordan's ruling compelled the companies to produce a new schedule for May 31st that "will respect a total of 175 days of hearing". An interesting way to observe World No Tobacco Day.

Mr. Duch's testimony will continue tomorrow.