Rethinking AIDS Jan./Feb. 1994

RETHINKING: We're interviewing Robert W. Maver, FSA, MAA. Bob Maver is one of the founders of our Group. You come from the corporate world, don't you?

MAVER: That's correct. My position was Vice President and Group Actuary for a major insurance company.

RETHINKING: For those of us who don't know this obscure but very impressive field, what is an actuary, and what kind of training is required to make one?

MAVER: The actuary is a relatively small profession, although important to the insurance industry. Actuaries are the ones who do the statistical work-up and statistical background to project when certain events will happen, and how often they will happen-for example, the probability of becoming disabled, or the probability of dying, the probability of living being the opposite to that. We're involved in designing policies, designing insurance products, and most importantly, pricing them.

RETHINKING: Let me clarify that. An actuary calculates risks for death rates for different purposes. Is that correct?

MAVER: Yes, and we have to relate everything to the financial world. We're very concerned with the present value of the future risk.

RETHINKING: From what I understand, a CPA takes about two or three years of undergraduate training, and then he has to pass a battery of tests that takes, I think, two full weekends or so. What is the training of an actuary like-just for beginners, entry level?

MAVER: It's somewhat rigorous. We go through a series of ten exams to attain a professional designation of FSA; that stands for Fellow in the Society of Actuaries. The ten exams start with rather traditional mathematics; for example, the first exam is a calculus exam, the second is probability and statistics, the third is a numerical analysis and the theory of interest exam. Then we get into the more esoteric mathematics of the insurance industry, where again we are combining the present-value concepts that the financial world is familiar with-the present value of future interest rates, for example-but we combine it with what we call life contingencies. the probability of living or dying in any given year. I would say that the actuary takes probably eight years on average, maybe longer, to pass this series of exams.

RETHINKING: If I were a CDC epidemiologist, how many years of training would I have, and would it be as rigorous as an actuary's training?

MAVER: You would probably have a Ph.D. at this point, if you were one of the top epidemiologist at CDC, so your background might be a bit more focused, specifically in the area of epidemiology, of course, and the statistics that go into that. But I would say you would have a four-year undergraduate degree, just as an actuary would; mine happens to be in applied mathematics. But I would say the CDC epidemiologist would have another four years of education, of course, to get to the Ph.D. level.

RETHINKING: And actuaries do this arithmetic, calculations, complex computer data base analysis, everything, with the idea in mind to commit premium dollars for various risk pools. That's it - it has to be a very fine probability calculation in the end, isn't that right?

MAVER: That's correct.

RETHINKING: So you're certainly in the world. Now, professionally, you're one of these actuaries, and on top of that your corporate credentials have led you to be the head of the entire department of actuaries for Mutual Benefit Life. And Mutual Benefit Life is one of the top ten in the country in size?

MAVER: At the time that I worked for them, they were one of the top fifteen. They were fourteen billion dollar in assets.

RETHINKING: And essentially the actuarial department was able to decide the allocation of premium dollars. The executives used that information to come up with a conclusion, is that right?

MAVER: Yes, and my specific area of responsibility was the group insurance aspect, with which listeners would be familiar. That's the benefits that they get from their employers.

RETHINKING: Now, you're sitting around here as an executive of sorts, and along comes AIDS. And you have to do what all good actuaries do. You've got to go out and look at the numbers and decide what the insurance company's risk is, to underwrite these things, or even to take the hit that might happen. Is that correct?

MAVER: That's an excellent description. It really caught our attention in the mid-1980s.

RETHINKING: And you were getting projections-at the time the information you were getting was that we were going to have a million deaths in a five- or ten-year span, and your department was responsible for going out there and finding out what was really happening, and what the risk was, and what the people who were currently on your insurance policies were going to cost you. Is that right?

MAVER: That's correct.

RETHINKING: Now, this is where you got involved with this whole thing, isn't that right? You began to look at the numbers. Why don't you take a few minutes and just tell us what the sequence of events was, so our non-expert listeners can get an idea of what you ran into.

MAVER: The Actuarial Society puts out various models to help the actuary in practice in a company. They put out these models to help you project for your own company what the impact is going to be of AIDS claims into the future. Every model that I looked at suggested that we had a major, major catastrophe, a major epidemic on our hands-one that was going to spread well beyond the initial risk population; one that should cause us to reexamine all of our underwriting rules; and one that really painted quite a gloomy picture for a group actuary, in the sense that one might have to make rate adjustments immediately, meaning premium increases, in order to prepare for such an epidemic.

RETHINKING: Let me stop you here just briefly. That means that the implications were that insurance companies had to do one of two things-either increase their reserves and cut into their profits, or increase their premiums and drive their customers away-if this was all going to come to pass, or even risk bankruptcy if they weren't properly capitalized, if they had too many AIDS claims. Is that correct?

MAVER: Yes. Well, especially along the lines of the point of view of group insurance. The nature of a group insurance contract is that you get to renew it every year. You get to set what you think is the correct rate each year into the future. It's not a lifetime contract. So we were actually looking at decisions such as, "Are there areas of the country where we can no longer write certain products, because of the expected persuasiveness?"

RETHINKING: And this had an unexpected political side effect too, didn't it, with regard to the gay community?

MAVER: Oh, absolutely, absolutely. Initially of course that's where the epidemic was when we started looking, and one has to be very careful of all sorts of insurance-based laws designed to protect our policyholders. There are certain kinds of underwriting that you can and cannot do, to better define the risk. But back to your original question. Essentially, what I found when I examined these models was that the data that we had so far was not at all consistent with the models that I was looking at. That is, the models would predict a number of AIDS cases for the year 1988, and I'd be able to look in 1989 to see how good that model was, and quite frankly. the model was awful.

RETHINKING: Awful meaning off by ten percent, five percent? What sort of expectation should a good model have?

MAVER: Well, I'll answer the question this way. The model was off by more than fifty percent. We don't have to get into the fine gradations here, of what's a good model and what isn't. We know that's a bad model.

RETHINKING: That's a bad model; there's no doubt about it. But if you found a variation of five or six percent, it wouldn't necessarily be a bad model, would it?

MAVER: No, no, that would have been fine.

RETHINKING: But we're talking about fifty percent. Now, who designed the models? Was it actuaries, or was it the CDC, or was it the data that was provided that you built it around?

MAVER: The data came from CDC. In some cases it was actuaries who took that data and tried to extrapolate from it, project out to the future. However, there were certain basic assumptions that actuaries were not well equipped to challenge, shall we say. I'll give you an example of a critical assumption that I found when I looked through the models-and of course what you do when you come up with a model that's not producing the reality is look and see what assumptions were made, and is it conceivable that some of these assumptions are in fact incorrect? Maybe there are places where they don't even realize they have made assumptions. The first thing I noticed, the assumption that all of the models used as far as future AIDS claims, was that 50% of people who had HIV, the virus that is alleged to cause AIDS, would convert to AIDS cases within a ten-year period. That is a critical, critical assumption-anyone that had HIV, not just within a certain risk group, anyone with HIV is going to develop AIDS within ten years. I decided to look into what that was based on. Surely there must have been a population that was studied to come up with that sort of assumption, and indeed there was. However, the population that was studied was a population from San Francisco that all had in common hepatitis B.

RETHINKING: No kidding! The projections were made on a very narrow, specific population.

MAVER: Yes. As a matter of fact, it was a population of gay men who had the hepatitis B, had various venereal diseases, had cytomegalovirus, Epstein-Barr virus, a whole host of problems in addition to HIV; let's put it that way. And the immediate question that came to my mind was, is this a reasonable model for the population at large that may contract HIV, or is this a model for a population that clearly has many, many other risks?

RETHINKING: So what's the next step, after you've figured this out?

MAVER: My next step was to gain some education in the medical arena, as to what were the reasons that we decided that HIV caused AIDS. That was another assumption that to me was made rather quickly. I noticed that regarding the group that was studied, of men with HIV who also went on to develop AIDS within ten years, one could ask the question, how about all these other viruses that were also present? Why did we decide it was HIV? This turned out to be the critical question, because this led me into quite an interesting series of papers and meetings, and pretty much pointed to the conclusion that it was a hypothesis, and a tenuous one at best, to suggest that HIV would always lead to AIDS.

RETHINKING: Was it at this time that you found out about the Group, or helped to form the Group-that you met Duesberg and heard of the other dissenters?

MAVER: Yes, as a matter of fact the first paper that I read was Peter Duesberg's in Cancer Research, back in-I think he published that in 1987. There were so many points that made sense in terms of keeping an open mind toward questioning whether HIV is in fact the cause of AIDS.

RETHINKING: Let me fast forward a bit. The net effect in terms of profitability, in terms of the company you were working for-what did that do? Was it able to reduce the panic in the boardroom and change your reserves and so on? Can you describe that process? Or was it so controversial that you actually ran afoul of some of your colleagues?

MAVER: Well, it certainly is a controversial notion to suggest that HIV is not the cause. However, it's not controversial at all to suggest that HIV is only a small portion of the picture of AIDS. I guess the essence of my research was to dig into the CDC data base, to actually go into the computer records, where they list, for every AIDS case that's ever been recorded in the United States with the CDC, perhaps 50 elements of data describing that case. What I was able to do from that was to reassure the company that in fact the epidemic is very, very strictly contained within certain high-risk groups, especially groups related to drug abuse.

RETHINKING: Let me interrupt here, now. Peter Duesberg takes the position that biologically speaking, the drugs can cause the damage. Now you're statistically correlating drug abuse with AIDS, is that correct?

MAVER: No question about it.

RETHINKING: So we have two different approaches now to verify, or at least to indicate drug abuse. By drug abuse, does that mean a specific type of drug abuse. or is it any long-term drug abuse? Is it a high correlation? Is it one to one, or what is

MAVER: Well, in the data that I looked at from CDC, they record intravenous drug abuse. What I was able to uncover, digging into the data itself, was that the vast, vast majority of those cases characterized as heterosexual AIDS by the CDC are actually in affiliation with the intravenous drug abuse in some way.

RETHINKING: I see. That's a little sneaky of them not to report it that way. Or did they not know it? Did they actually claim ignorance about the drug relation?

MAVER: I guess that's hard to answer.

RETHINKING: Well, let's skip the political side. Now, you found this out; your company was able to make its adjustments. Were you the first company to do that, or did the entire insurance industry figure this out pretty much at the same time?

MAVER: I think that the industry is still operating on the premise that HIV is equal to AIDS, and that even though the vast majority of the population with HIV have not gone on to develop AIDS, the insurance industry at large believes that they will go on to develop AIDS-that HIV is the equivalent.

RETHINKING: So that's where it stands now. Did your insurance company at that time take that posture, or did they actually lessen their reserves, or whatever you call it, to take advantage of this new information you had developed?

MAVER: No, it did not lessen reserves. It only used the information to understand better the nature of the risk that we were underwriting.

RETHINKING: There's a political element there too, of not wanting to inflame the gay community, or those who thought it was judgmental to know these things. Is that a possibility, that was part of the equation at the executive level?

MAVER: Well, certainly one has to be careful.

RETHINKING: In closing, I want to ask you this: how about the next two years? What's it going to take to get the truth out there, and do you have any hope for some kind of change, or is it just too far gone at this point?

MAVER: Well, I still hold out hope for change, and I think that it will occur by doing good science. I think that's the only way out at this point. I think we have to find an organization courageous enough to do the studies that should have been done many, many years ago. As an example, there are some viable theories-one of which is Peter Duesberg's, one of which is Bob Root-Bernstein' s-as to what may cause AIDS. These are theories that would be testable in an animal model, and I would hope that we would move forward with those animal models and have those tests done, so that we can either rule out those theories or confirm that in fact, yes, there it is.

RETHINKING: Some of our subscribers who will be listening to this tape are HIV positive, and really don't have any other kind of health problem, nothing-really that's it for them, and they're scared. Given that that's entirely true, what chance do they have of developing anything remotely resembling AIDS, just from having that virus and nothing else?

MAVER: From the research that I have done, it looks to me like it's virtually impossible. They would be the first case ever on the books of having HIV only.

RETHINKING: That is really a fantastic comfort, I think, to people who are afraid of this, and whose doctors would like to give them AZT prophylactically. Thank you, Robert Maver.*