The Washington Post published a poll on the special election in Massachusetts last week, as part of the Washington Post-Kaiser-Harvard poll series. The poll carried obvious implications for the health care reform debate (as did the election). And like other polls in the series, which are frequently health care-related, it was co-directed by Harvard professor Robert Blendon.
Blendon sits on the board of Assurant, an insurance company. As I’ve noted previously, Blendon consistently fails to disclose the affiliation in his healthcare polling work or in his various bios (here, here, here). It is so hard to find this affiliation noted anywhere — even his lengthy Harvard bio leaves it out — that I get the sense that Blendon feels he has something to hide.
Since Assurant has an obvious and substantial interest in healthcare reform, and Blendon’s polling work for the Washington Post frequently concerns healthcare, shouldn’t the Washington Post notify its readers of this conflict?
Blendon has been around the health care reform block before. During the Clinton healthcare reform effort, he produced a number of healthcare-related polls for Harvard and the Robert Wood Johnson Foundation (his former employer) that played an important role in gauging public opinion on the issue of reform. He was invited to the White House to give presentations and was frequently quoted in the press as an expert observer.
At that time, as he is now, Blendon was conflicted. He joined the board of the insurance company Fortis, Assurant’s predecessor, in March of 1993, just as the Clinton healthcare reform process got underway. And then, as now, Blendon was identified as a Harvard professor, with no mention of his ties to the insurance industry. Regardless of whether these ties influence his work (that’s impossible to determine and besides the point) the timing of his role at Assurant and the lack of disclosure looks very bad.
Fortis’s health insurance subsidiary, Time Insurance (now Assurant Health), was also part of a coalition of businesses supportive of the Clinton plan’s employer mandate. Time Insurance’s CEO, Carl Schramm, participated in at least one meeting at the White House, like Blendon. Schramm arrived at Fortis/Time Insurance from the Health Insurance Association of America (HIAA) one month before Blendon joined its board, in February 1993.
In December 1992, in preparation for the Clinton healthcare reform proposal, Schramm and the HIAA had produced its own version of a healthcare reform proposal. It called for “a new Federal law that would require coverage for all Americans, define the basic set of benefits, and try to contain health care costs by limiting tax breaks for the purchase of insurance” (read up on the plan at Wonk Room).
At the time, Schramm said the plan was “the only way you preserve the private health insurance industry. It’s plain-out enlightened self-interest.” A few months later, Schramm and Blendon joined Fortis/Time.
HIAA went on to become AHIP, the powerful insurance industry association which has played a prominent role in the current healthcare reform process.
I wouldn’t have noticed Blendon’s most recent poll if I hadn’t read this post by Nate Silver. Silver observed that the Massachusetts voters surveyed by the WaPo special election poll were actually strongly supportive of the state’s universal health insurance law, despite their opposition to Obama/Coakley.
Blendon’s polls reliably produce positive results on this question, which he has polled on at least three times in the last six months. He then, in turn, is cited favorably by influential healthcare reform supporters like Paul Krugman and Nate Silver, who find his poll results agreeable (they are generally supportive of Massachusetts-style reforms).
Ironically, in his post on the Blendon poll, Silver says that he is “reluctant to cite polling conducted by activist groups.” It seems reasonable to be cautious about the agenda that these groups bring to their research (assuming this is Silver’s reasoning), but shouldn’t similar levels of caution be applied to polls conducted by industry groups and insiders? Silver brings considerable analytical rigor to his work; he would do well to apply some of this rigor when evaluating the origins of the polls he uses.
Simply because a source hails from an academic context, should it be trusted as free of a conflicting agenda? Of course not. Blendon is a perfect example of why this is the case.