Felix Moore
Year: August 8th, 1989
Location: Granlibakken, California
Interviewed by: Blackburn, Henry
Abstract
Felix Moore, pioneer CVD statistician-epidemiologist, headed statistics at the National Heart Institute at its origins and devised the grid for the Framingham recruitment. He went on to nurture epidemiological activities of the early American Heart Association. He provided fine insights, very openly, about the origins of CVD epidemiology, about Framingham and government service, about the AHA Pooling Project that he managed, and about the ‘Gofman Inquisition’ (a different viewpoint).
Sound stuff is here on the beginnings of Framingham, which complements Gerald Oppenheimer’s historic analysis from the Bethesda correspondence. Moore introduces positively the innovations of Jilson Maeders, the first head of the study. He clarifies the recruitment of the Framingham sample. He offers his outlook on the contentious scientific debates that took place behind the scenes, including the hostility to funding for Framingham, the working relationships with John Gofman in the Cooperative Lipid Program and in the American Heart Association Pooling Project. There is much about the early Heart Institute and the National Health Survey, and there are anecdotes throughout. (KR/HB)
Quotes
The first institute was the National Institutes of Health and then largely with Mary Lasker’s doing, there was organized the National Cancer Institute. And the third was the National Heart Institute whereby NIH now became the National Institutes of Health. This was …. I’ve forgotten when the Cancer Institute was organized, probably about ’46. It was just after the War, at any rate, with Scheele as the Head, who subsequently became Surgeon General, of course, largely promoted by Mary Lasker who was always interested in more and more statistics. The main idea was to use statistics to get money. So the Heart Institute was organized in 1948, late in the fall.
CJ Van Slyke was the first Director. And Harold Dorn, at that time, was the Chief Statistician for the Cancer Institute and informally, the Chief Statistician for NIH. Although I don’t think there ever was such a position actually created. Van Slyke, I don’t know who told him to do it or whose idea it was, but he went to Harold Dorn, quite properly, for a recommendation. And Harold recommended me for the Heart Institute.
So he called me for an interview and I think one of the first things I said to him (it was a one-on-one interview) is, “In all honesty I’ve got to tell you that I don’t know a heart from a valentine.” And this was literally true. I had no background in biological science at all. Which I’ve always regretted. But Van said, “Well, Harold Dorn says you’re a good statistician and we’ll teach you cardiology.” Which he proceeded to do. He was not a cardiologist himself, either. He had other people on the staff and I was given a lot of time just to deal with them and to read and so on. I’ve packed in, I guess, a lot of learning. . (10)
Moore’s Contributions:
There are three things I have done that I’m very proud of. Framingham, of course, and the Atomic Bomb Study reorganization in 1955, which I did with Tommy Francis and Seymour Jablon. The Framingham Study, of course, first with Jillson Maedors and then with Roy Dawber. . . And the National Health Survey. Which became a National Health Survey not until 1957. . . I slip that in because I think that’s been important for epidemiology. Mine [contribution] was solely with the interview part. So those are my three. (11)
Framingham Origins: “The very important person there is Joe Mountin. I doubt if either of you knew him. But Joe is way up there as Deputy or Assistant Surgeon General. A great public health innovator and so on. And he had the notion of setting up community heart disease programs. As it turned out, there were two of them set up in the Boston area affiliated with Dave Rutstein’s Department of Preventive Medicine. One was headed by a guy named Robbins in Newton. It was sort of a heart disease control demonstration project. And the other one was set up in Framingham under the directorship of a very young doctor, Jillson Maedors. Whose name is on, I guess, the first publication that came out. But for reasons which are not a necessary part of the story, Jillson was replaced by Roy Dawber.”
A tough start:
But that [first article in the AJPH in 1956] tells the history starting with MacKenzie and it says what we intended to do. And, by god, that’s what we did. Because of the shift and the lapse of time in bringing in a new director and everything, we lost almost a year’s headway at Framingham. And it wasn’t a total disaster. We tried to rescue it, we made some compromises which in retrospect I wouldn’t have done. But you couldn’t throw away all the work that had been done.
Now Maedors had the right idea. He was going to get the community which was away from Boston, but where he could have the prestigious Boston cardiologists help front the project. Which they did magnificently, Paul Dudley White and Howard Sprague, Sam Levine, … and I’ll come back to them if we have time. (12)
More Framingham:
They were gung ho. They were going to get the whole community in back of this thing. The first thing that they did was to get the President of Dennison Company and family to come for examinations. Dennison, you know, made all those little pieces of paper. They were the best known family in town. There were only two other industries – one a GE assembly plant, and the other Westclox which was not actually in Framingham, but right on the fringe.
Then every person who came in was to bring in 10 more persons, and you know the game – the old pyramid or chain letter type of thing. People just flocked in. I think Jillson always had some notion that maybe he’d get around and do that sometime (recruit from a grid). But I don’t think he ever would have. This thing was being so successful. When I first came into contact with this, they had literally almost a year’s backlog of reservations or appointments. And you know, this is just clear sailing.
In the first place, the decision had to be made whether the Heart Institute would take this over. I went up there with Van Slyke and Jim Shannon and we spent two or three days there and when we came back I wrote a memorandum recommending to Van Slyke that the Heart Institute take over. (14)
Framingham Planning: What I did was impose the grid. Framingham, like many of the other Massachusetts towns, is a central built up area, a township or a small county, in western terms. They had a census of inhabitants which they got out every year. It was a register of voters, but it was a census of inhabitants, it had names, and addresses, and ages. So I had in hand, and at the time of the 1950 census I had this checked out by the Census Bureau. They assured me that this thing was about 97% complete. At the Census Bureau I couldn’t see the original returns, of course, because those are confidential. But I could give them the book and say would you match it against your records. So we had that assurance, although that was late in coming. But I had other assurances that this wasn’t quite good on a house-to-house canvas basis.
So, figuring that we could handle about 6,000 people in the study, and there were 10,000 people of the appropriate ages, and we decided on 30 to 59. I simply imposed this on what had been done already and said, “We’ll draw the sample from the list, and then anybody who’s already signed up is in the sample. Those who’ve we committed to we’ll go ahead and give examinations to. The rest of them we’ll try to get them in.” So we drew 6,600 names on the theory, although we thought of this as a study of disease among males, we figured, and this is where the health educator came in, I think she was right. That if we got the women to come in, they’d drag the men along. And I think that pretty well worked out. In other words, we used that kind of pressure.
And for political reasons, if we drew anyone from a household within the eligible ages of 30 to 59 as of 1950, then we would draw everybody else in the household who was of eligible age. So we would get the husband or the wife or anybody else by the same name. In other words, relatives of the husband. So it was ad hoc sort of sampling. I don’t think anybody ever tried doing it exactly the same way. We essentially guessed at a 1% per annum incidence of diagnosable coronary heart disease (based on the men) and figured that in 10 years that would give us enough for some kind of analysis. But we set out 20 years as the target. (16)
Poor Cousin from the Outset
Now Jim Shannon, as far as I know, never objected to that memorandum [about a 20-year plan for Framingham]. And again, I don’t know how you want to handle this. I think Jim Shannon is a great guy, but he’s no epidemiologist. And I think Jim Shannon always felt when he was Clinical Director in the Heart Institute, that every nickel that was being spent on Framingham was being taken away from his labs. He either had to be in the laboratory or at the bedside. I’m sorry to say that Berliner continued exactly the same. I went and talked with Berliner personally after I was long out of the Heart Institute. I was on either two or three different committees. Everyone unanimously recommended the continuation of Framingham and just as unanimously, they’d had a big fight back at Building 1. to save the money, you know. Still fighting, that’s right. So they had to go out and get grant money and piece it together here and there. Frederickson was a little more favorable because he wanted to get the blood samples on a community basis and there was something in it for him. Some of the others I thought were friendly, others. . . . (18)
Framingham Analysis:
The other thing is I organized the processing of the data. I did all the analyses in this first report and one other one which Tavia Gordon is principal author of and deserves a great deal of credit for. Which was a sort of methodological analysis of response and so on. Subsequently, Tavia decided that there was not enough difference between what we called our SX sample and the sample that is the volunteers. So he merged those with the original sample. I had no objection to that. There was a certain amount of self selection, we knew that. People who are going to die very soon of heart disease just stayed home. People who moved away were, in general, and we looked at those who moved between the first exam and the second exam, they were generally healthier than those who stayed. And I think this is inevitable in any sort of epidemiologic study. So I never felt we needed to apologize for the fact that we didn’t get anything like – we got 66%. But then that loss includes all the people who had died before we could get to them and/or who had moved away before we could get to them. And because of this long delay in getting through, it took us four years to get around the first sample, we had much larger losses from those reasons than we would have had if it had been an ordinary. . We thought we could cover this in two years. And essentially it’s covered on a two-year cycle as you know now. (20)
Framingham Publications:
On the other hand, it’s true we were accumulating the data on the follow-up end and [felt] it wasn’t so essential. Joe Mountin kept writing me, “When are you going to be publishing, when are you going to publish?” And I said, “I am not willing to publish the prevalence data because I don’t think we can talk about what kind of a sample it is. But I think that the incidence data will be the proof of this and I’m not so concerned what the baseline is, what kind of a sample it is if you’ve got good follow-up of the population which looks like its kind of generalizable.” (39)
London Transport Study:
“I simply, from my own memory traveling in London, asked Jerry [Morris] not completely in all innocence, that I thought I’d remarked the difference between the bus drivers and the conductors, who obviously had quite different patterns of physical activity and also size. He later on told me that he had gone back and discovered that they had records on belt sizes upon issuance of the uniform and, indeed, the bus drivers were different from the conductors.”(21)
Origins of a new AHA Council for Epidemiology:
Katz. He just cut my legs out from under me. You see a motion for tabling is not debatable. So I couldn’t even get a chance to present our position. And I guess I showed the fact that I was really indignant because I was. I wanted them to know that this was not something that gentlemen ought to do to each other. I was told that his motives were good and there may have been others similarly motivated, but I never heard this from any other source, that the old-line councils felt that they were being weakened by having their clientele taken away. On the other hand, he never showed any animosity that I could see toward epidemiology. . . And afterwards I went to Oley personally and maybe somebody else went with me, I don’t know. Oley had been already President of the Association. . . But I said, “Oley, I can’t do it. I don’t have any prestige with these people. You do and if we’re going to have a Council, I’d like to have you take over the committee and maybe they’ll listen to you next time, where they wouldn’t listen to me.” (29)
The AHA National Pooling Project
And so we became a Council and I proposed, and I probably had proposed this before we became a Council, the notion of a Pooling Project. But the first step seemed to be this Committee on Criteria and Standards. Which I, frankly, was disappointed in the end result. We had staff men on it and I don’t know. It may have been my fault because I was Chairman for a period of time. And as Chairman, I became almost a permanent member of the Council’s Executive Committee. Which I think was Marge Bellows doing. The rules and regulations weren’t very tight. I just always found myself being a member at the next meeting. Eventually Fred took over the Committee Chairmanship. But I retained the Chairmanship of the Pooling Project. Which I guess, literally speaking, may have been a subcommittee of the Committee on Criteria and Standards. But, again my permanent membership on the Board Committee or whatever, seemed to continue on and on. . .
I remember one time we had a little trouble getting the paper [Pooling Project results] on the [AHA Sessions] agenda. . . I got Fred to give the history and Kannel to sort of summarize. I read most of the paper, yeah. Then I figured….. You remember about the publication don’t you? This thing batted back and forth…. I wrote a first draft. There’s a three-man steering committee, editorial committee. Doyle who did not do one single bit of work. I wrote the first draft and I admit it was a hurried job but I had all the statistical parts of it together and Jerry [Stamler] agreed to write the final draft and did a very good job. Finally the thing got done and we sent it off to the Journal of Chronic Disease and they committed themselves to publish it. And the d….d printing plant burned down.
At any rate, that delayed it for another five or six months. But I have to, I guess, take a lot of the blame for it being delayed so long. And I know the committee got pretty teed off about the continued delays. We had some internal problems. Computer problems. I remember Rick [Shekelle] helped a great deal. He ran the logistic until we were able to come out with almost identical findings. And it was only then that I was really comfortable. I have every confidence in Jerry Cornfield, who I think is really one of the great statisticians of our time. I think if I had to say one thing, it really did establish that a group of essentially, prima donnas, that you always find in medical research, could come together and stick together for that long and put out, what I think, was a fairly decent report. [. . . not just decent, Felix, it’s the foundation for everything that happened after. I mean, every time somebody wants to reference the assertion that blood pressure, serum cholesterol, and cigarette smoking are related to the incidence of coronary heart disease, it’s the 1969 pool. . . . For instance, you were talking about obesity. Well, Elizabeth Barrett Connor within the last two or three years, undertook a very large review of that and her taking off point for that is the Pooling Project. (31)
The Gofman Study: Cooperative Lipid Study.
I guess as much as anybody, I started it. Because I didn’t believe the guy. I literally didn’t. You couldn’t reason with him. He was a boy wonder who became a boy wonder emeritus. You know like ….. you refer to Leonard Bernstein – a boy wonder emeritus. I guess they used to refer to Hutchins in Chicago the same way. He came to the Heart Council [of the National Heart Institute] asking for a whole lot of money to establish….His stuff was just too pat. You couldn’t predict with any biological measure that I knew of as accurate as his atherogenic index.
So the Council decided to support a cooperative study and they signed up [Fred] Stare, who didn’t believe in it either; Max Lauffer in biophysics at Pittsburgh and I’ve forgotten the younger man who was with him; Framingham, Albany, and Gofman’s own collection of data , , , I’m talking about the groups that contributed the serum samples. The serum samples, both the cholesterol and the SF [ultracentrifugal] values were done in the local labs, except for Framingham where they were sent to Gofman. So one thing we did have, at least on the cholesterol, we had Gofman’s cholesterol and Framingham’s cholesterol. At that was Tavia’s idea. And I think one of the great ideas in this business. I don’t think Tavia’s gotten sufficient credit. You think so? Ok.
That ended up, as you know, in a shouting match. The fact of the matter is, and there’s still a few people alive who can verify it, that it came down to a point that Gofman threatened to sue me, personally, for distorting the data. Now, somewhere along the line, we got all the things in, and Gofman wouldn’t accept our report. . . A minority report was appendix A and the majority report appendix B. . . . I had, and god knows how I had the idea, but I had the idea that we might be in for trouble. I got the Committee to set up a review committee and I asked Paul White if he would chair it with Sprague and Levine. I personally presented all of the evidence that we had on the new events to that committee in a series of meetings. I sat there and heard them discuss this. Is this definite? Is this probable? Is this questionable? [for coronary endpoints]. . . by agreement, they did not have access to any of the data on the lipids. . . .The three best cardiologists in the world. And by god, I was happy I had done that. But I just cut Gofman’s legs out right from under him.
John issued this threat and Stare and Page, who easily out-ranked John in prestige, came and stood behind me and we wrote the majority report which was report B. We got the last word. That wasn’t the end of that but. . . So then we published the normal values afterwards without him, which is probably the biggest set of cholesterol data that had appeared up to that time. That was the end of it. But I must say that in retrospect I figured I misspent about three years of my life on that. Because there were some things I should have been doing at Framingham which I wasn’t doing. (35)
Get in there somewhere my relationship with Van Slyke. One thing he told me right off the bat when we started this so-called Gofman Study. Never, never trust a laboratory. And this came out of his old VD days. And he said, “The first thing you’ve got to do is establish that these people can measure what they say they are measuring.” And the results were horrible, horrible. (36)
Full Transcript Access
Full transcripts of interviews may be made available to those engaged with original materials for scholarly studies by contacting us.