Dr. David Bell

David Bell

MILLER: This is Dr. Bess Miller, and I'm here with Dr. David Bell. Today's date is August 30, 2017, and we are in Atlanta, Georgia, at the Centers for Disease Control and Prevention [CDC]. I am interviewing Dr. Bell as part of the Oral History Project: The Early Years of AIDS, CDC's Response to a Historic Epidemic. We are here to discuss your experience during the early years of CDC's work on what would become known as AIDS [acquired immunodeficiency syndrome]. Dr. Bell, do I have your permission to interview you and to record this interview?

BELL: Yes, it's a pleasure to be here.

MILLER: For this oral history of AIDS at CDC, we have been focusing primarily on the early years, beginning in June 1981 with the publication of the first MMWR [Morbidity and Mortality Weekly Report] on the five cases of Pneumocystis carinii pneumonia among homosexual men. David, you were one of the principal epidemiologists at CDC to work on the infection control aspects of HIV [human 00:01:00immunodeficiency virus]. For the area of HIV infection control, focusing on risk of transmission and strategies for preventing transmission, I guess the early years continued through the '80s and into the '90s. Starting in 1987, when there were limited data on risk of HIV transmission and healthcare workers were afraid to care for patients with HIV infection, you provided leadership in this area and served as chief of the HIV Infections Branch, once this branch was formed, in the Hospital Infections Program in the National Center for Infectious Diseases [NCID] until 1997.

Of course, you then went on to provide leadership in many areas, including antimicrobial resistance, influenza, and prevention of mother-to-child transmission of HIV, collaborating with the World Health Organization [WHO] and 00:02:00other agencies. You also provided leadership in emergency response activities, most recently including serving as the scientific lead for CDC's Ebola virus response in 2014 and 2015.

Let's begin with your background. Would you tell me about where you grew up, your early family life, and then where you went off to college?

BELL: I grew up in Philadelphia. It was inside the city. It was in a neighborhood with many immigrants, primarily Ukrainians from Eastern Europe. I went to a large public high school of 4,500 students in the next neighborhood over, which was a German neighborhood. Most students in that high school did not 00:03:00go to college. In fact, the overwhelming majority did not. But my parents favored and impressed on me the value of education, and I was lucky enough to be good at my studies and worked hard. I got a scholarship to Princeton, where I went to school for undergraduate training.

MILLER: Who or what influenced you to go to medical school, and where did you end up going to medical school?

BELL: I was always interested in science. In college I majored in physics, but 00:04:00physics was taking too much time. I liked other topics, humanities. I actually spent a year in between junior and senior year in college going to school in France and studied philosophy. When I came back, I decided I wanted to be more involved with people than in a physics laboratory. I also took a biology course, which was fascinating. It just shifted me off to medical school---- it seemed like I could actually do something useful and meet lots of different kinds of people. I was lucky enough to go to Harvard Medical School, which was a completely wonderful experience for me. I enjoyed [it] much more than college, actually.

MILLER: What did you decide to do for residency after medical school?

BELL: I did a pediatric residency at Boston Children's Hospital. Then I came to CDC from 1979 to '81 in the Epidemic Intelligence Service [EIS] program here. Incidentally, I remember being one of the earliest people in the loop in June 1981, because I was in viral diseases and responsible for cytomegalovirus, which was cultured from the initial patients. I said, this is a very common virus, we don't know what this means--which, of course, was true. So I heard of AIDS at the beginning--

MILLER: How did you decide to go to the Epidemic Intelligence Service?

BELL: Alexander Langmuir, as you know, is kind of God at CDC. Things are named 00:06:00after him, and he's quoted reverently. He was the director of the infectious disease block during my second year of medical school. This was unprecedented at the time. It was usually a microbiologist who was director of the block. He assembled lectures on different topics. But he would always--for every bug we learned about, he would describe an outbreak. There were actually three people in my medical school class who went to EIS, plus there were a lot of EIS alumni around who just had wonderful experiences to share. It was great.

MILLER: Who were you assigned to? You mentioned viral diseases. What were you assigned to and what did you work on during EIS?

BELL: From '79 to '81, I was assigned to a variety of things, but my main area was a mysterious illness called Kawasaki disease, which was described by Dr. Kawasaki in Japan. It's a disease of young children: fever, rash and a variety of other symptoms. About 1% of them actually developed heart attacks due to coronary aneurisms, and the cause of it was completely unknown, still is. During my EIS time, there were two large outbreaks in the United States, and there had never been outbreaks--well, at least not in the United States. So I got to work 00:08:00on that, and it was quite interesting. We didn't find the cause, but we at least ruled out a lot of things, which was helpful at the time. Then I went to do an infectious disease fellowship at the University of Rochester, which was the site of one of the outbreaks. So I discovered I liked the program there and those people. But eventually, [in] 1987--

MILLER: Why did you come back to CDC?

BELL: After infectious disease fellowship, I had an academic job, which I'd always aspired to, but I decided I didn't like that. I didn't really like academic medicine. We didn't know what to do. [We had] just bought a house, had 00:09:00another baby on the way, and we weren't just going to move for no reason. So I worked as a general pediatrician for three years. That actually was enormously helpful later, because I understood then what it was like to be the end user of CDC recommendations that would oftentimes tend to get written in--you know, for better or worse, somewhat of an ethereal environment. But I knew, what am I going to tell a parent actually? So that was quite helpful. In 1987, I got called by a former colleague at CDC and asked if I would come back and work on AIDS in the Hospital Infections program, to be the AIDS coordinator.

MILLER: Who was that?

BELL: That was Dr. [William R.] Bill Jarvis. The reason this all happened was, 00:10:00there was an uproar in 1987, a major national uproar. Three healthcare workers had gotten HIV infection, not from needlesticks or sharp instrument injuries contaminated with blood--that was previously documented. That was known to unfortunately be occurring at a steady trickle. But this uproar was [because] three healthcare workers got HIV infection from getting blood on their skin or what we call mucous membranes --like on their mouth and eyes. And these were also, I'll just say, outside the main AIDS cities of New York and San Francisco. One of them was in a Midwestern state, they didn't have much AIDS. This 00:11:00particular healthcare worker just was holding a gauze pad at the site of an arterial puncture. This was in an emergency where a patient came in to be resuscitated and died, and she [the healthcare worker] was holding a gauze pad and it was discovered, died of AIDS. She was documented to have gotten infected as a result of this. There was another person who was filling up a phlebotomy tube. The top fell off and blood splattered in her face. There was another one who was fixing some tubing that was clogged up and blood----

So the combination of, you know, you can get this awful disease from just 00:12:00holding a gauze pad with your finger, plus the fact that this was now throughout the country, AIDS--this is the mid '80s. The background for this was that the cases were increasing with no end in sight, were spreading throughout the country, outside places like New York and San Francisco, and people were starting to get worried about it. Well, there was an uproar, and so I was recruited to come back and take the lead in dealing with this.

MILLER: Big assignment. So by then there were about 50,000 patients reported with AIDS. This was not the days [of being] reported with HIV infection but the 00:13:00disease, and the majority of these people were dying.

BELL: Yes. That's the other thing that's important to know. This was before any antiretroviral drugs. This was a death sentence, and it was a horrible death, a slow horrible death. So the idea that you could get this from holding a gauze pad was just amazing.

MILLER: CDC had done a case-control study looking at what were the risk factors.

BELL: No, not yet. That was later. We did that. Let me tell you. So the question was, when I got back, what is the risk to healthcare workers and what can we do 00:14:00to reduce this risk? There was very little known about this. For one thing, it was a very multifaceted question. There are a lot of different kinds of healthcare settings and healthcare workers. There's outpatient, inpatient, emergency department, surgery, obstetrics, the wards, the laundry room. We were getting laundry workers with needlesticks because they found a syringe in the sheets that were sent to the laundry. There are all different settings.

What was the risk of getting infected, from which fluids, what about the cumulative risk? None of this was known. And then how to reduce these exposures? To prevent them? Now, we didn't know what the risk was. We had some information 00:15:00to suggest that it wasn't huge. There were AIDS surveillance data that showed some number of cases of AIDS in healthcare workers who said they had no other risk, but they didn't have documented exposures. They were mostly men. We knew healthcare workers were mostly women. None of them were surgeons or anybody who you would think might get stuck.

MILLER: Was there a thought that perhaps these people were men who had sex with men or drug users?

BELL: Absolutely, and just weren't reporting it. We did, however, have from AIDS surveillance--it was a handful, but it was worrisome-- a steady trickle of very well-documented situations where there was a documented exposure to HIV-infected blood, needlestick, whatever it was, the person tested negative then 00:16:00seroconverted. It was very well documented. Again, it was a handful, but it was increasing and the epidemic was spreading, and we were worried. One of the biggest fears at the time was that these reports might have been the tip-of-the-iceberg type phenomenon. HIV infection, you know, can go for years with no symptoms.

At the time the other thing I have to say [is that] blood exposures were so common they were thought to be unavoidable. I mean, this was normal. When I went to medical school, well, for centuries before that, you get some blood on you, so what? You wash it off. Even a needlestick was a nuisance. We had no idea how 00:17:00many needlesticks occurred, and they were thought to be unpreventable, thought to be unavoidable. So our job was to find out the risk, the frequency of these, and what are the circumstances that increase and decrease the risk, and how do you prevent them. One of the big concerns at the beginning, as I mentioned, was the tip-of-the-iceberg effect. Oh, I forgot. There's one other thing we did know, which was the risk from a single needlestick exposure of getting infected with HIV, exposure to HIV-infected blood, was 3 in 1,000.

MILLER: How did you know that?

BELL: Because CDC had started early, around 1982 or '83, collaborating with a 00:18:00couple hundred hospitals around the country to prospectively enroll workers who reported these exposures, tested negative, had documented exposure, and later see how many seroconverted. We knew that the risk was 0.3% --and actually there were smaller studies in place at San Francisco General Hospital and at the NIH [National Institutes of Health] Clinical Center that also found the same.

MILLER: So once those got going you could actually measure seroconversion.

BELL: No, no. Actually it was before that.

MILLER: Oh. How did they do that?

BELL: It's a really good question, because these were patients with AIDS. A healthcare worker got stuck, and in the beginning all we knew was the patient had AIDS and the healthcare worker was just followed to see whether they got 00:19:00sick or not. I mean, that's the best we could do in the early '80s. Now, in 1985 when the antibody test became available, the patients could be tested and then the workers could be tested. So basically, we didn't think the risk was huge, but there was a risk, 0.3%.

I'll just say one other thing: even how that risk was described was polemical. We got criticized for describing it as high-- or low (because it wasn't as high as hepatitis B, where a similar exposure might be 30%), -but on the other hand, a 3 in 1,000 risk of a fatal disease is not exactly low. There were people who 00:20:00accused us of concealing the risk to healthcare workers. The government, CDC is concealing the risk to healthcare workers because they don't want healthcare workers to be afraid to take care of the patients. There was one actually very loud, outspoken orthopedic surgeon, and Congressional hearings [about] what the government is hiding from you, CDC is concealing the risk to us. Then there were people who weren't that extreme, [who] were concerned--some healthcare worker unions in particular, that we were understating the risk, which could lead to a false sense of security. Then there were people on the other side, who were concerned that there might be a perception of exaggerated risk. These tended to be AIDS activist groups, also people in healthcare who had to manage situations. 00:21:00There were situations where---- we had dietary workers who would leave trays of food on the floor outside the patient's room. They were afraid to even enter the room, much less draw blood or something like that. So they were very concerned about how we even described the risk that we couldn't quantify that well.

Anyway, in this environment after these three--we called them the splash cases, that was the moniker--one of them wasn't a splash, it was just holding the gauze pad---- but anyway, what CDC did was recommend "universal precautions:" the blood and certain body fluids of all patients [was] to be considered infectious and precautions should be taken to avoid contact. So gloves [had to be worn] to 00:22:00avoid contact with blood, also goggles and masks.

MILLER: Tell me a little bit about how that was delivered. Was that a recommendation and report in the MMWR?

BELL: Yes, there was a big consultants meeting, getting expert advice from----

MILLER: Who would be at such a meeting?

BELL: Maybe 60 infection control experts from around the country, from stakeholder groups who had important perspectives to contribute, public health officials with some experience in managing a recommendation at a local level, for example. There were a lot of questions. Should all patients be tested when they enter the hospital? The outcome was basically that wouldn't really solve the problem, for a couple of reasons. One was [that] a lot of these [exposures] 00:23:00happened in emergency departments where [patients] came in and needed to be resuscitated--there was no way, there was no rapid test in those days. Also, it would only be as good as the most recent test result anyway. So the decision was universal [precautions].

However, many people thought blood exposures were inevitable. Particularly the surgeons had a problem with this [recommendation] ---- the orthopedic surgeons, because in truth, to prevent sharps injuries, we didn't have a lot to say. We said things like, be careful and don't recap needles by hand because you might stick yourself, and dispose of them properly. But the surgeons said, look, we've got bone fragments that we're operating on, we've got bone saws that aerosolize blood. We are careful, but we can't----Anyway, so the orthopedic surgeons--there 00:24:00was one particularly outspoken [who said that] the government is concealing and all this stuff. So to address the tip-of-the-iceberg phenomenon, we actually started with the orthopedic surgeons.

MILLER: Before you get there, was this the first recommendation of universal blood and body fluid precautions? Which sounds kind of like a watershed event in clinical practice, and [that] they didn't have that for hepatitis was of a concern. This was a huge --

BELL: This was a watershed in 1987. Actually before that, there had been a category of infection control precautions called blood and body fluid precautions. For example, patients with hepatitis, if they were known, were on blood and body fluid precautions. In 1982, CDC recommended that patients with 00:25:00AIDS and other conditions, if it seemed like they were going to become AIDS, blood and body fluid precautions apply.

MILLER: How did universal precautions differ from what was in '82 the blood and body fluid precautions?

BELL: All patients, regardless of whether they're known to have infection transmittable by blood or not. All patients. Actually, in 1985 CDC had come out with a recommendation for that, but it wasn't well publicized and didn't get much attention --and this was in 1987 with the splash cases. This was front-page headlines: what are we going to do about this? All blood is infectious. That was everybody. Then, of course, it was how to implement that --and these exposures were happening anyway, particularly the needlesticks.

So there were several categories. One was the risk, and then there was the exposure prevention. I'll get to that in a minute, but the tip of the iceberg--I have to tell you about this because it's just such an amazing event that happened. The American Academy of Orthopedic Surgeons, God bless 'em, this was a courageous move. They agreed to cosponsor with us an anonymous serosurvey at an annual meeting. Now, it took three years to get this arranged, to get all the permissions from that organization. Also up in Washington, we had to get permissions. I mean this was an era when a lot of people were very afraid about what a study might show, and so sometimes they were reluctant to let it be done. But eventually--

MILLER: What might it show? That there is homosexuality--?

BELL: Well, no. But let's just say that there was this huge iceberg. Let's say that 25% of orthopedic surgeons were infected. That would have been a complete nightmare for all healthcare. There would have been pressures to have them disclose who was [infected]. All healthcare workers would have said, "you see?" But on the other hand, the unknown is ultimately worse. You have to punch through this and get the answer, because you don't know where to target your prevention efforts if you don't know what the risks are. So anyway, what happened was that the annual meeting was in Disneyland, California, in the Disneyland Convention Center. We had this enormous white tent in the parking 00:28:00lot. In the tent were dozens of anonymous counseling and testing stations, booths. It was all anonymous. Surgeons would come in--it would actually be quiet during the meeting, and then during breaks there'd be waves of them. We'd have to tell them, turn over your name badge, turn over your badge. So they got anonymous counseling, testing, they got phlebotomy, linked by bar code on the tube, and they filled out a questionnaire about risk.

MILLER: What did the counseling--how long was that counseling session and what did it involve? At that time there was nothing much to offer people.

BELL: This was the recommended CDC counseling and testing. Before anybody gets counseled, you have to be sure they understand what they're doing here and what 00:29:00HIV is and what the results mean, what options are available for you. Even if you test negative, if you're engaging in risk behavior, you should stop doing that, because just because you tested negative now doesn't mean you are not at risk. Then once you test positive, to get into care.

MILLER: In '87, was there any care?

BELL: Actually, the study didn't happen until '91. In '87, no, there wasn't, but it took years to get permission. But it had to be anonymous. You see, there was an innovation here. Normally counseling and testing -- in local health departments, the counselor knows your results. But here, we didn't want anybody to know. The innovation was, they got counseled for both positive and negative results. They didn't have to reveal their result. The counselor said, if your 00:30:00result is positive----if your result is negative---- And the testing, normally testing with the confirmatory testing in those days took time. There was a local laboratory, and the tests were done around the clock under supervision of CDC lab people who went out there to supervise this. Tubes of blood would be drawn, and the next day go out with bar code results. The whole thing was organized bedlam. I mean, people running by with walkie-talkies. In those days we had walkie-talkies to communicate with each other at the other side of the tent, for example, and phone calls from Atlanta and Washington [to ask] how's it going? There'd be reporters who were denied entrance to the tent. Some of them pitched a fit, but no. This is where we could--the American Academy of Orthopedic 00:31:00Surgeons said, this is our tent. If it had been the government, it would have been harder. They said, no, it's a private organization.

Anyway, at the end there was something like 3,400--in 4 or 5 days, 3,400 surgeons got anonymous counseling and testing. It was about half of the surgeons there. Nothing like this had ever happened. Only two were positive, and both on their anonymous questionnaire had reported other risk factors. So people breathed huge sighs of relief. It didn't mean they weren't at risk and that there wasn't still work to be done here, but it did a lot to dispel this tip of the iceberg [concern], and then we could get down to work. Okay, what are the known events and the real risks, without having to worry that there was all this. It was totally amazing. I give them a lot of credit, because it took 00:32:00courage. It took that organization [the American Academy of Orthopedic Surgeons] courage to do it.

MILLER: How did you manage it? Did you have a group of people in your own shop that helped oversee this?

BELL: Yes. It was a joint project. There were people in our branch. The director of it was an epidemiologist, Dr. Mary Chamberland, and there were also other people [working] very closely. There were lab people, but it was our--it was a good thing.

MILLER: Sometimes program and lab don't work as well together, but it sounds like this one did. Was the lab--

BELL: It was [Charles A.] Charlie Schable. He was the serology--I see you smiling, because he was a legend. He was in charge of the AIDS serology lab at 00:33:00CDC. He was the one who went out there, and his people made sure that that local lab did it right.

MILLER: Were there any state involvement in that, or was it directly federal and professional organizations?

BELL: I recall somebody from California State. Certainly they were informed about all this. But, no, it was really us, because these people had come from all over the country. The convention happened to be in California, but it wasn't a California thing. We needed to talk about exposure prevention --there's the 00:34:00risk, then there's exposure prevention. So we started--exposures occur in all different settings--so we implemented observation studies in operating rooms, obstetrical departments, emergency departments, around the country.

MILLER: What is that?

BELL: We had observers watching a surgical procedure, recording various information. Not the patient's name or anything like that---this was all anonymous --but recording when surgeons got stuck. Or people in emergency departments, they'd walk around and watch things happening.

MILLER: Is this a standard hospital infections procedure for other diseases, was 00:35:00it then ----

BELL: No.

MILLER: What are the risky procedures?

BELL: Back then, there was no objective quantification of risky procedures. You had people anecdotally saying, "I get stuck all the time," and there was no quantification of this. There was no description of the circumstances, the risk factors, to get some rates and get some idea of how often and where do we target interventions. So this was extremely helpful. We did these studies in the big city hospitals, we did them in suburban hospitals, we did them both inside and outside the high AIDS cities, you know, places like New York and Los Angeles, to 00:36:00try and get some idea. A study like this had also been done in San Francisco General Hospital. This was Dr. Julie Gerberding --back then when she worked there, but that hospital had a lot of experience dealing with AIDS patients. We needed to get a sense more broadly about how often these exposures occurred around the country in different kinds of hospitals.

MILLER: How would that work if, let's say, you were the one going to observe a surgical procedure? Would you have a checklist of things you want to look for?

BELL: Yes. We gave them a checklist to fill out, and if something happened, describe this. If it was a hand, which finger, nondominant hand, or just what happened. Another thing that turned out to be really important later when we 00:37:00started talking about risk to patients--and I know we're going to get to that-- how often after the surgeon got this injury did the needle re-contact the patient? In other words, what was the potential for exposure of the patient to the surgeon's blood? Some of this was their [the patient's] bone, it was already in the patient--or the surgeon would just keep on suturing [after getting stuck] and say, ah, and keep on suturing.

We actually found that about a third of the time, in a third of these injuries, that somehow the patient was also exposed. Now this became important later, when we started talking about risk to patients, but certainly at this time it was, 00:38:00how do we prevent this? There were technique issues. It turned out that certain procedures were riskier, certain techniques--if the surgeon held the tissue being sutured with his fingers as opposed to an instrument, we could demonstrate the increased risk. Some of this is in retrospect, of course. At the time these techniques were historical. Nobody really cared that much, and it took like, no, you need to--the surgical societies, I will say, when we published this result, they jumped on this right away. This doesn't happen anymore, these things. These procedures have really gotten a lot safer.

But the biggest problem, really the most intractable, that wasn't very susceptible to observation, were the needlesticks that occurred during 00:39:00phlebotomy -- getting blood from a vein or an artery. This occurred all over the hospital. Many different doctors, nurses, medical students would do these things, the clinical lab people, and you really couldn't observe. So we had to try to rely on reports. The documented occupational infections were actually not coming from surgeons. They were coming from phlebotomy, and so this was the real problem. The insight into how to deal with this came from an injury epidemiologist at the University of Virginia, Dr. Janine Jagger. This was a couple of years later. She had the insight from that perspective that these 00:40:00syringe needle devices were inherently unsafe -- and also [that they] were used for many tasks that you didn't need a needle for. For example, what would happen was, somebody would draw blood into a syringe from a patient's vein or artery and then she--I said "she", these were actually almost all women--would then transfer the blood from the syringe into a glass tube--tubes were glass--and they had rubber stoppers, and so she would have to apply pressure to get through the stopper with this needle in this syringe full of blood to transfer the blood into the tube. And, yes, sometimes she would miss and jab herself or squeeze too hard and the tube would break. These--we had a steady-growing stream of these 00:41:00[cases]--this was how people were getting infected.

With this insight, the unions picked up on this, and OSHA (Occupational Safety and Health Administration) and Congress even eventually passed laws requiring "engineering controls," was the phrase, safer needle devices. So nowadays, to transfer blood, it's plastic. You don't use a needle to transfer blood between tubes anymore. I mean, that's ridiculous. Or we used to --for example, holding pieces of tubing together, you'd stick a needle to connect a tube. This is ridiculous. Also the syringe--either the needle is recessed, or the needle that actually gets into the patient-- there are devices now, for example, they have a 00:42:00sliding sleeve that you push from behind forward---- it covers up the needle tip and locks into place, so this needle is now disabled. It's never going to stick anybody anymore. That's how really those problems were solved, but that took a while.

MILLER: So CDC it sounds like did quite a few studies, as well as academicians and San Francisco General Hospital and elsewhere. Was that a natural flow? How was it decided what studies CDC would take on and--

BELL: Let me just say we worked collaboratively--we depended on people in 00:43:00hospitals to actually do--we gave them funding to do this, but we coordinated, organized the study and had designed the protocol, and then we were blessed with wonderful people to collaborate with us. It became a joint venture. When I say CDC, I don't mean it was just us here, sitting here in Atlanta. There were people out in hospitals who were actually getting this done, and it was very interactive.

MILLER: How did that work in terms of collaboration and managing and----you must have had periodic meetings.

BELL: Sure. The multicenter studies were complicated. We might have half a dozen hospitals doing one of these observation studies, and they had to follow common protocol, but it worked. We had dedicated, conscientious people at work who very 00:44:00much wanted the answers to these questions, just like we did. They got support from their institutions to go ahead and do it.

MILLER: Would you have, for example, in your branch or division project officers for each of these studies, who would go out----

BELL: Yes, and just make sure things were going right and everybody was informed. We'd bring the investigators from these hospitals to meetings in Atlanta where we'd all share experiences and plan the next improved study. Okay, now what did we learn, and what's the next step. It was all very interactive.

MILLER: In some aspects of the AIDS early years, there was a lot of competition among high-level academicians in terms of getting those first New England 00:45:00Journal articles out. Did you experience much of that in this arena, because this was pretty hot stuff?

BELL: It was hot stuff, but it wasn't--there weren't that many people doing it. Occasionally an issue would arise, but in general----in AIDS research-- this wasn't considered hot. The hot stuff was the virus, the community. This [our work] was of huge interest to hospitals and healthcare workers, and it was hot for them, but there wasn't--I don't rule out that there might have been somebody [who] wanted to publish something, but that wasn't a huge problem for us. There 00:46:00weren't enough of us to compete with each other, really.

MILLER: Also CDC, in terms of trying to get at risk and prevalence of infection, did some seroprevalence surveys. That was under your unit?

BELL: Right. The biggest one was the orthopedic surgeons one. There were also some smaller ones, Dental Society [American Dental Association] annual meetings, that were somewhat along that line.

The real other problem, if you'll allow me to jump ahead a little, was the issue of, okay, once somebody does get exposed, then what do you do? We call this postexposure management.

MILLER: So the years we're talking about now are still the late '80s, before there's a real solid antiretroviral regimen. Now we had AZT [Azidothymidine].

BELL: Yes. Even in the mid '80s--you still had to know what to tell the person, how long to follow up. The antibody test became available in '85, so how long after [the exposure] do you need to get tested, and what should you do in the meantime. The source patient should get tested with consent. What if he didn't consent? What should you tell the worker? It became clear that most seroconversions, when they occurred, happened within the first six months, actually the first three months, but then there would be reports of delayed --like nine months, very well documented. Does everybody get followed for a 00:48:00year? This was hugely distressing for the exposed workers, as you can imagine. There was no cure for this fatal disease. To say your risk is only 3 in 1,000-- you know, if it's you--and we'd say, during this follow-up period, [have] safe sex, don't donate blood. Safe sex, that means we're telling women not to get pregnant. Most of these people were women in childbearing age, so if you tell one, okay, for six months don't pregnant, then she gets another needlestick, okay, another six months, don't get pregnant. Sometimes they already were pregnant, which was totally terrifying for them in terms of the baby. So there were lots of gut-wrenching stories here.

AZT, it was called Zidovudine then, was licensed in 1987. The NIH clinical 00:49:00center decided they were going to make it available to their workers. Nobody knew if it worked or not for postexposure. NIH decided they were going to make it available anyway, even though they didn't know any better than we did if it worked. So as a result of that, questions came from around the country: well, if the NIH is doing it, shouldn't we do it, too? Well, but there are side effects. We didn't know if it worked or not. There were issues about how to implement that---- [it] has to be given very soon. So we did have a meeting and came out with recommendations, I guess in the late '80s or something like that, about---- we don't know if it works or not, but if you're going to use it after exposure, here are some considerations: these are the toxicities, the doses, this is what 00:50:00you should be watching for.

But the question was starting to burn: does it work post-exposure? Because, remember, there were other exposures, too. We were getting questions that tended to come to us about rapes and sexual exposures. We had to find a way to assess this and, you know, the advantage--"advantage" is sad--of a needlestick injury is that you know exactly what happened, when it happened, many details of the exposure, the source patient, [and] you can follow [the exposed worker], you can document. For sexual exposures you're relying on a history. You don't know really what happened or how many times or who else or anything like this. These [occupational] exposures and infections were continuing to occur, so we just had to figure out a way to assess this.

Now, the study design that you're supposed to do is to take workers who get exposures that come in and report, and then you randomize them. You give half of them the drug Zidovudine, you give half of them placebo, and then you follow them for a year or whatever and compare the percentages of who acquired this awful fatal infection, and see if there's a statistical difference. You know, the Burroughs Wellcome Company, the maker of the drug, they tried that. They tried. But the problem was that a healthcare worker who got a needlestick didn't want to be randomized. They wanted the drug.

MILLER: In terms of ethical review board, would somebody pass that--would that type of study designed have passed?

BELL: Yes, because nobody knew if it worked or not. You know, it could have made them more likely--I mean, you don't know. When you don't know, you don't know. So it passed, but--[there were] two problems. The healthcare workers didn't want to be randomized, but also, since the baseline risk was only 3 in 1,000, in order to demonstrate what we call statistically significant reduction in that risk, we would have needed thousands of healthcare workers-- just from the pure math statistics, to do the comparison. The company tried, and after a year, they 00:53:00only got 84 workers willing to enroll, as opposed to the necessary thousands. So this wasn't going to work. We had to figure out something else to do. What we eventually figured out was----this isn't kosher, but there was no other way, and there was a compelling need to figure out some way. We took reports of occupationally infected healthcare workers reported from AIDS surveillance. Well documented--, whatever it was, needlestick, etc., test negative, source patient. Well-documented cases, these were the cases. A couple of them were in our prospective surveillance projects in these couple hundred hospitals of 00:54:00healthcare workers who got exposed and then followed, but most of them were just documented at the time of the needlestick. They reported to occupational health, so they were well-documented cases. The controls came from what we called the needlestick study. The study had been going on since about 1982 to see what happens if they had well-documented exposure, and most of them didn't get infected. So taking the cases and controls from two different datasets and two different methodologies is really not [ideal]--but there was nothing else to do. We had great statisticians who did the best they could to make sure that there weren't some biases creeping in. They teased out what's called confounding 00:55:00variables, Because, the more serious, in the mind of the healthcare worker exposed, the more likely they may have been to report. Statistically it was very well done.

MILLER: Who was the statistician?

BELL: Dr. David Culver was in the Hospital Infections Program. The person who was the lead for this study actually we recruited later, but she took over the study. We recruited her from Brazil. She's an American citizen and also Brazilian, and she now is the Chief, it was Dr. Denise Cardo that became the director of this division. It's one of the best things I ever did at CDC, was to get her to come and work for us in our branch.

Anyway, this study took a while to do, and the results were published in, I 00:56:00guess the tail end of '95. It was during the government shutdown, but the MMWR published this. There was a statement from the CDC director that the public health urgency of these results was such that despite government shutdown we have decided to publish this, even though we're shut down. They were published in the New England Journal of Medicine a year or so later, maybe a year and a half. Basically, the comparison ---with needlesticks--- the comparison of healthcare workers that got infected- versus [those who] didn't get infected, basically it was a larger volume of the source patient's blood involved. The 00:57:00source patient was closer to death, which at the time --particularly these data from the early '80s, the best we had --we thought that this represented high viral titer. They were close to death, [with] no treatment. So this makes sense: more blood from the source patient, high viral titer, sicker patient, and very convincingly what reduced the risk: if they got AZT, the drug, post exposure, [the risk was] reduced by 89%, [which is] statistically significant. We had a consultants meeting, and they all said, this is it. So we then issued public 00:58:00health service recommendations. This came out in, I guess, early '96: what types of exposures were riskiest, for what types of exposures should drugs be recommended, versus offered, versus not recommended. Also not just this drug anymore, but also the whole load of the newest drugs. This zidovudine is an old drug. We're using the newest ones now, including this old one, in case there's some reason only it works, but we're going to add the others for the most serious exposures.

MILLER: You mentioned that that came out in '96 and the data from the original needlestick study were ready in '85, something like that.

BELL: Well, it was ongoing. I mean, this is a surveillance project called a study----you know, these were monitoring people who got exposed.

MILLER: That's a long, long time. What were you recommending to people during all of those years?

BELL: In terms of post exposure prophylaxis, we did have a statement, like I mentioned, in the late '80s, maybe it was early 1990, but Zidovudine was available in '87 and the NIH started offering it maybe in '88 or '89. In '90 we said, we don't know if it works or not, but if you use it, here's how to use it. So that went along for a while.

MILLER: I'm reminded of ACT-UP [AIDS Coalition to Unleash Power], and I don't 01:00:00know, whether union 1199 for the healthcare workers and so on. How did you interface with them during this period of, we're not sure if it works but maybe it works? Is the drug widely available? Were you a part of that?

BELL: They were totally supportive---on an issue like this, which was really a medical scientific issue, not so much a political issue. The drug was available. We weren't saying don't use it. We were just saying, we don't know if it works, so we can't recommend it.

MILLER: Was that an issue of controversy, because we know there was a lot of controversy over why is the government not providing more AIDS drugs and people are dying. So here you were sitting on--it was an issue, I think, that CDC has 01:01:00frequently had to deal with, which is we don't have quite enough data but we have a great demand.

BELL: But this was very different. These people didn't have--their risk of getting infected was 3 in 1000. So these weren't like people with AIDS dying because they didn't have access to the latest experimental drugs. These were people who the overwhelming likelihood was they were not going to get infected, and just if you wanted to reduce that risk further, you could take this drug, which was by that time widely available. We had other issues with these advocacy groups that I'll get to in a minute. But on this particular one, this wasn't----

MILLER: Wasn't an issue.

BELL: Yes.

MILLER: Again, there's a long time period we're discussing here, but do you have a sense about how many healthcare workers were infected during that decade? Are we talking hundreds?

BELL: No, no, no. Dozens, including around the world. We developed close collaborations with colleagues, particularly in France and England and Italy, who were wrestling with these problems, too, and the well-documented cases--my memory slips--I mean it probably got into the hundreds after a while, but this was by the late '90s.

Now, they would have had to have been documented and reported. This was the problem. People have to go right after the injury, test negative, then they have to get followed. This all has to happen. That was actually one of the concerns that I was talking about that tip of the iceberg, you know, a disease where you might not report your blood exposure and you don't have symptoms anyway. So the true number is probably higher than that. I'm not even talking about places like Africa where the infection control issues are completely different, but it was enough to be worried about.

The interesting thing, I found very interesting, is that after this somewhat 01:04:00unorthodox methodology for the study that found efficacy of zidovudine post exposure after needlesticks, a year or so later the Public Health Service recommended this for nonoccupational exposures, too, for sexual exposures. Then it became recommended by WHO and there's never been another study like we did, for a combination of reasons: [It was] impossible, maybe unethical, so it's this study from needlesticks that is the basis for all the post-sexual recommendations---- it's sort of interesting.

People came to us for information about--we've been talking a lot about blood, but what about the other body fluids? We had exposure risk information on other 01:05:00body fluids like urine, and people would be asking about saliva and sweat. So in a sense, even though our focus was healthcare workers, the information we gathered was useful to colleagues in dealing with other questions, too. And we learned from them, of course, so it was a great team. I know you were part of it there.

AIDS in the '80s it was----when I came back in 1987, there were lots of legal 01:06:00issues and confidentiality. I know we're going to get in a minute to the risk of the patients from the dentist, where this all kind of exploded, but when I first came back in 1987, we had an issue, some legal issue, and I had a meeting in the head CDC lawyer's office, Gene Matthews. I walked in, my boss (Dr. William) Bill Martone introduced me, and Gene looked at me and said, "Welcome to the pit." And that's right, it was--there were some days it felt like combat. We were getting ranked out from people who said we were taking instructions from the gay lobby, and then there were others who were [saying] we were kowtowing to conservatives in Washington. You know, we were trying to just put one foot ahead of the other and try and get more data. I've not seen this. I've been at CDC now for 32 01:07:00years, and there was nothing like it. You can't even describe it to people. It's hard enough to deal with scientific uncertainties about a virus and what you should do about a virus. And sometimes there are political issues like in pandemic influenza --who's going to get the countermeasures--and the border control. But there was nothing like AIDS, where people were scared and had competing versions of morality even, sometimes.

MILLER: Were you dealing with a lot of stigma against gays and drug users, in terms of trying to assess the problem or trying to document things? What were some of the causes for this sturm und drang that you are describing?

BELL: People were afraid, with very legitimate fears. I'm sorry if I implied that these were somehow inherently unreasonable people. But people were afraid. Gay people had a lot of infected people in their community. First of all, they were afraid of losing their loved ones, dying. They were afraid that they would become ostracized, and when they needed medical care, they wouldn't get it. They saw that. I told you about the dietician [who] would leave the food [outside the room]--so they were afraid. We had other people on the other--I don't mean to 01:09:00say other side, but a lot of times it sort of--who were also afraid of blood, of doing their job, going to work and getting infected on the job. And what's the CDC doing about this? If they even just worried that we were not doing something right, or doing something that might exacerbate the things they were worried about, I mean they let us know, and it wasn't always very polite. The politicians would get involved, too, and start shouting at us.

MILLER: How was that? Either federal or state politicians--this was the Reagan era and a little bit after that. What were some of the tensions you got there?

BELL: The issue about testing or not would come up----but in order to really get the full flavor of that question, I think it would be better to move to the situation where the dentist transmitted to the patients, because that's when the public really got alarmed and the politicians followed them.

MILLER: Let's move to that watershed case. You were involved in a lot of watershed issues, the so-called Florida Dentist case. Can you introduce us to the case itself?

BELL: Yes, this was in 1990. The Florida Health Department reported through AIDS 01:11:00surveillance a young woman who later identified herself as Kimberly Bergalis, who was about 21, who had AIDS. She was interviewed by experienced investigators, and they couldn't find any risks. I mean, she did not acknowledge any risks. It just turned out that they also happened to know that her dentist also had AIDS, because this was also in Florida. So they sent the HIV strains from the patient and the dentist to CDC in Atlanta for the molecular virologists 01:12:00to analyze them and compare them. There was a relatively novel technique they had of sequencing and comparing, and they found that they were essentially the same strain and that this couldn't have been--it's unlikely it would be coincidence, couldn't be, at 99 point percent-whatever. So the question was--this first became known, I guess it was around June 1990. So what are we going to do?

I remember going to a meeting with [Dr. James W.] Jim Curran, it was [in] a small crowded room, you know, what to make of this? People went over the results and over everything and there was just this--people were like quiet and then shouting and then quiet again. The realization was sinking in that she might 01:13:00have gotten infected from the dentist. She just went to get her wisdom teeth out, and what could this mean? So what to do next? CDC had to publish this quickly, because really there was no additional information we were going to get any time soon that was going to change this. We had experienced investigators, we had the HIV strains, and then we had to have a process to lead the discussion as to what are we going to recommend. And boy, was that a firestorm. I mean, it turned out actually that five patients were infected in that practice, not just one. But the problem was that the dentist died and the practice was closed, so 01:14:00in terms of an investigation as to how this happened, that became difficult.

[Dr.] Harold Jaffe was really the one who directed--he and his colleagues, [Dr.] Carol Ciesielski and some others--directed the investigation. My role was the development of the recommendations, like what are we going to do about this? So immediately the two sides--there was the public uproar. I mean, you can imagine. This young woman, she died. She died. From the dentist! There were people who said that all healthcare workers should be tested and positive ones should not 01:15:00be allowed to practice, at least not practice any type of invasive procedures. On the other hand, there were people-- these were AIDS groups and public health groups --who were concerned about all this attention and resources devoted to this issue. They felt the risk was negligible, and there was a very high risk that these infected healthcare workers, who were mostly gays as a matter of fact, would be discriminated against. They felt like all we should do was to just say, you know, reinforce infection control precautions and that's all you need to do. Then the politicians started getting involved. The Senate actually passed a bill providing jail terms for infected healthcare workers who did 01:16:00invasive procedures without getting patient consent.

MILLER: Did it pass?

BELL: It passed the Senate, yes. It died in conference or something, but I mean this was huge. We actually were quite convinced that for the vast majority of healthcare workers, even if they were infected, there was no risk they would transmit to patients. From what we knew, we were convinced. Even the ones who did invasive procedures, we were convinced that for the vast majority, the risk was negligible. Yes, they maybe needed to be sure about their techniques. But we had reason to believe that the risk wasn't negligible in every situation, and so 01:17:00we were not willing to just say, oh, just follow recommended procedures and that's all you need to do. We knew the limitations of those recommendations.

We also had information that-- There's another blood-borne virus, hepatitis B (HBV), which is present in the blood at much higher levels than HIV. So it's not the same risk, but there were many well-documented situations where HBV-infected surgeons had transmitted to patients. The risk was higher during certain types of procedures. There were no observed violations in infection control, and 01:18:00sometimes the guy [healthcare provider] wore double gloves. There was a super-spreader phenomenon. Some didn't transmit to any patients---- some transmitted to multiple patients. A lot [was] not understood --it was enough to give us pause in terms of just saying, oh------ We also knew from our observational studies that I mentioned earlier that, after a surgeon got injured in the operating room or stuck with a bone fragment or something, a third of the 88 of these [cases] witnessed by our trained observer, [he] witnessed the needle re-contacting the patient after sticking the surgeon, and this happened including attending surgeons. It wasn't just trainees. So we were not willing to 01:19:00just say, oh-okay--. So what did we say?

MILLER: Were you surprised that this emerged from a dental practice?

BELL: Not hugely.

MILLER: Was that even considered during looking at observational studies?

BELL: No, not really. We were not hugely surprised, to be honest with you. There was a group of dental officers at CDC. There was a dental unit epidemiologist, and we did have the sense that--as in many outpatient settings--I want to take care not to stigmatize dentists here, but many outpatient settings were just 01:20:00freestanding operations. There was no hospital infection control practitioner to go around and make sure things were done right and the latest recommendations were followed. Talking about a private office somewhere, there's nobody. Dentists were all in this category. We actually were not enormously astonished, let's put it this way, that this might have happened in a dental office.

People, based on their fears, legitimate fears, they wanted the outcome of our recommendations to be one way or the other. They did everything they could to try to accuse us, influence us, including people who I would have to say should 01:21:00have known better --who were our more traditional colleagues and friends. We got pressures, let's just say pressures, to either just say nah, no problem, or to restrict all of them. We tried to figure out some kind of way forward here, and I'll tell you what we did. This went through more than a hundred drafts. It took a year.

We ended up with recommendations that clearly exonerated the vast majority of healthcare [workers]. We're not talking about phlebotomists here or even 01:22:00invasive procedures, but we said--this didn't work in the end---, but let me tell you what we tried to do. We tried to say, based on the information we have about hepatitis B and other studies, certain procedures we'll just call "exposure-prone." I'll just mention one of the procedures that was highly associated with hepatitis B transmission from a surgeon to patients, and also in observational studies was highly associated with injuries, and then re-contact with the surgeon's blood, was a procedure called vaginal hysterectomy. So the hysterectomy done by the vaginal route--this was before laparoscopic surgery. (I mean the problem also went away. I'll tell you about that in a minute.) But the surgeon would feel where the tip of the suture needle was with his finger. He 01:23:00[or she] couldn't see it. It was blind suturing. We said, certain procedures we'll call "exposure-prone----" professional societies should designate which they are, and people who do those procedures should voluntarily get themselves tested -- voluntarily. If they're positive, they should seek advice from an expert review committee that involved somebody who knew about the procedures they did and their medical condition and so on. If the expert review committee said it was okay, then it was okay. Or maybe they shouldn't do certain procedures, or something like that.

That took a hundred drafts and a year, and in fact it was all printed in the 01:24:00MMWR ready to mail out---- a hundred thousand copies--I don't know, whatever. Then the instruction came down from Washington: you're going to add a sentence that [says] the patient has to consent to this, regardless. Even if the expert review committee says it's okay or just don't do a particular procedure, the patient has to be informed that the surgeon or whoever is HIV positive and has to consent. This just blew away the house of cards we had tried to build up, because patients--whatever you think of the merits of that--- it did blow away our house of cards, because it meant that the surgeon's confidentiality was gone. If that patient knew he was positive, so would everybody else. It meant 01:25:00the whole thing was just not workable--so the hundred thousand or whatever MMWR's printed had to be shredded and new ones printed, adding that sentence that came. I was told who it came from, but I'll just say higher-ups in Washington. Wow. Then Congress did pass a law that all states have to adopt these CDC recommendations or they're going to lose their public health funding. That actually passed and was signed. So then we were left with the situation of these completely unworkable recommendations. Also, the professional societies declined to designate which procedures were exposure-prone. They said they 01:26:00couldn't do that.

MILLER: Why would that be?

BELL: They said it depended on the technique of the surgeon, and they couldn't do it. I think, yes, there's a lot of truth to that, but anyway---- Congress passed a law that all the states had to adopt unworkable recommendations or lose their public health funding. The states had to submit to us copies. The lawyers here just approved them all, except maybe if there was a confidentiality issue they said, no, you need to fix that.

MILLER: The lawyers here reviewed the state guidelines? Is that what you----?

BELL: Policies.

MILLER: Were the policies following the draft? What was then the CDC's position on what the policies needed to state?

BELL: There were a couple of elements that had to be mentioned, and confidentiality of patients and workers had to be assured. The bottom line was--

MILLER: But no, the patient was not to be informed--

BELL: No, no. And the states were stuck, too. They knew this was ---- so they put in some general stuff, and we just signed it.

MILLER: Did CDC ever come out with guidelines?

BELL: Revised guidelines?

MILLER: Mm-hmm.

BELL: That was considered. You know, this all happened during a Republican 01:28:00administration. It was actually Bush father in '90---- Clinton was '92, as I recall. So, of course, they [the Democrats] tend to be the more liberal people. The congressman from Greenwich Village was always hammering us about something---- how terrible these guidelines were. They were unworkable. When the Democrats came in, there was an expectation that this would be reversed. I was actually charged with drafting what it could say, how we would back away from 01:29:00this, and they decided to just leave it alone. They decided just to leave it.

MILLER: Was anything ever published in the MMWR?

BELL: Yes, the original guidelines --in 1991.

MILLER: But post Kimberly Bergalis?

BELL: Yes, this refers to the recommendations for prevention of HIV and hepatitis B transmission to patients during invasive procedures. This was the one that led to all this controversy. [MMWR July 12, 1991: 40(RR-08)----1-9.]

MILLER: But then it was rewritten, and how was--

BELL: It wasn't. We got early instructions to start doing this when the Democratic administration came in. Once they realized the firestorm on both sides of this issue, they just said, leave it. Plus the states had taken care of it. It was going away. Let me tell you why the issue was going away. For one 01:30:00thing, we did a bunch of look-back studies, of patients who had been treated by surgeons with AIDS or HIV infection, mostly AIDS---- but didn't find any more transmissions.

There was one more transmission found. This was in France, actually -- an orthopedic surgeon transmitted to a woman---- but there weren't any more [cases] found. Of course, they would have been hard to find---- that doesn't mean they didn't happen, because these people didn't have any risk and [wouldn't normally get tested]. Kimberly Bergalis developed AIDS a little earlier than usual, which was used to question her veracity by the people who didn't want to believe it, but [it was] still within the range of possibility. Well, some things happened. For one thing, the surgeons and the dentists, the professional societies, really 01:31:00cracked down in terms of fixing their techniques, "What do you mean, re-contacts?" They said, "Never again!" "You don't do these things, the blind suturing."

MILLER: Was it ever assessed what [was] the likely source of transmission in the dental office?

BELL: We couldn't find out. There were people--and this was another thing--there were people who wanted to believe that this was sloppy technique, or it was a sterilization or disinfection problem. The dentist had died, the office was closed, but the technicians--or maybe that he'd done it intentionally. There were all these people who just wanted desperately to believe that this didn't really happen -- or the reason was sloppy technique or even intentional. I guess in the end it can never be completely proven, but--the hygienist and the office 01:32:00staff wouldn't believe any----they said, no. One investigator said--we mentioned this to the hygienists, and they started laughing. They saw what was going on, and it was the recommended procedures. It was not egregiously sloppy or anything. But people wanted so much to believe this. We'd find experts in something, stating that this had happened [sloppy technique or inadequate disinfection/sterilization procedures], who knew nothing about the investigation. Nothing. But there would be luminaries out there, stating that they had inside sources, and then somebody else would say this---- and then they'd cite each other to confirm it! Some of these were, you know, normally our friends. They figured we were under pressure, which in the end it turned out we were under some. They were worried about the pressure from the right wing, but 01:33:00they also had their views, too.

Anyway, what was I going to say? Oh, yes, then when the administration changed and it was a Democratic administration, oh, finally, we'll get this all taken care of. Then the message came down, no, forget it. Oh, I know why it started to go away. The surgeons--I'll just quote-unquote say, they cleaned up their procedures ---and dentists [also]. Good antiretroviral therapy came along, triple therapy, so you could get an infected surgeon's or dentist's viral load down to undetectable. That took a lot of the risk and the concern away. Plus, malpractice insurance and hospital credentialing started to deal with this on their own. Surgeons and dentists actually started to voluntarily narrow their 01:34:00practices, because they didn't want to transmit either. So [it was] a combination of all these things---- the problem went away, but it was something.

MILLER: The HIV-infected healthcare workers started looking after themselves in terms of getting treated? How did it go away?

BELL: No, I mean this question as a public health issue went away. It subsided because there was better treatment. The surgeons paid more attention to their technique. The dentists particularly went to town on their outpatient infection control and education and getting people up to standard--and it did turn out 01:35:00that this [transmission] was really rare, based on all the subsequent studies.

MILLER: Fascinating, hard times.

BELL: It was very illuminating. The whole thing was--I guess I really took away--you know, I guess a few--we really need the data. We need to make our recommendations based on scientific data. It means we really need the data. It happened, surprisingly, a little too often that people were reluctant to see a study done because they were afraid of what the results might show, on either side. Well-meaning people, mind you. But you have to punch through that somehow, 01:36:00because if you don't have the data--of course, it has to be well-designed studies--then you really just listen to people shouting at each other.

Sometimes we have to make recommendations based on limited data, just like a clinician sometimes has to do something, even though he or she is not totally sure what the patient has. You can't just wait. Sometimes we have to do, and so you just get the best advice you can from experts, stakeholders, even the public, their perspective. Then you say, here's what we're recommending based on what we know. Then you try to get more data and say, we'll change the recommendations as we go along. But you need the data. You need to be able to change. I think the AIDS wars really demonstrated the importance of that.

MILLER: When you look back on that, how do you feel as you look back on those years and the work that you were doing?

BELL: Well, it was exhilarating. I don't think I could do it again. I mean, it was--but I knew why I went to work every morning. It was to try and help healthcare workers, patients, the country. I had good people to work with. CDC, the people I was lucky enough to work with, were wonderful. We had collaborators that I alluded to, experts, consultants. We had good leadership: Jim Curran, 01:38:00Harold Jaffe. I felt empowered. I felt like no matter what was coming down around me --that I might have to answer some question from a hostile audience, which happened fairly regularly, or draft something [controversial] or help or meet in the CDC director's office-- I felt like my own personal leadership and the leadership above was supportive. They were supporting our work, and I didn't have to worry about them. That was huge.

MILLER: Did you ever have to testify before Congress?

BELL: Yes, actually twice. That was also an experience. One in particular. There were a number of congressional hearings on these issues over the years. Of course, they tended to come at it from different directions, and I'd get a lot of coaching from the [CDC] Washington office and others here, to give me the background for this. One of them in particular, I remember it was chaired by [Senator Ronald Lee] Ron Wyden, a congressman at the time from Oregon, a 01:40:00Democrat. The genesis was unions were-- healthcare worker unions were quite active and interested in this and quite supportive, actually. Their concern was maybe we were understating things. But they had good people.

Anyway, there was a hearing on risks, but really the safer needle device type thing and what more could be done by the government, OSHA. That was what this hearing was about. It was amazing. There were eleven TV cameras in the hearing 01:41:00room. It started with a woman who'd gotten AIDS from an occupational exposure----testifying behind a screen so you couldn't see her face. She recounted the injury and the awful--I actually did happen to see her later. I got a peek of her as she was leaving the cafeteria. I didn't say a word, but she was just so gaunt and thin. It was heartbreaking. This happened to her on the job from a [needlestick]. Then there was a guy who said he was from the laundry room. He emptied onto the witness table a big bag of sharp instruments that he 01:42:00said were recovered in the laundry. I don't doubt it.

Then at one point--it was CDC, FDA [Food and Drug Administration] and OSHA. The three of us were together, and they said, now we'll hear from the administration witnesses. I was thinking, wait a minute, I'm a scientist here, I'm not the administration. But somebody from the Washington office said, no, no, that's the way they refer to you. Don't take it personally. Anyway, so we talked. It was pretty sympathetic. But this hearing really did get a lot of attention, and it spilled into other arenas. It was quite an experience.

MILLER: You sure have had a fascinating career. Did you ever worry that you 01:43:00would get AIDS during all of this, or that your family might?

BELL: We did have clinical--what's the word. I worked at Grady [Grady Memorial Hospital in Atlanta]. I did some clinical duties to keep in touch with reality, and I didn't actually worry because I knew how it was transmitted and how it wasn't.

MILLER: Anything you'd like to add before we finish?

BELL: I just think it's really a blessing. CDC is really a wonderful 01:44:00organization, and when you have good leadership, it can just do wonderful things. It is a government agency. We do have masters in Washington --- supervised by people the taxpayers elect to supervise us because we're spending their money. I don't know all the dynamics, but we do need to have enough independence to do what the scientific data seem to indicate to us. Sometimes I guess higher-ups--I spent a year at [Department of] Health and Human Services---they have a very different perspective. They have a ship to run up 01:45:00there in Washington. We're down here, the scientists, and they've got--they have to fight the political battles that I just mentioned and complained about. So they, no hard feelings here, we do have to be able to maintain some independence and credibility. I guess after spending a year working on those recommendations, to see them shredded for them to put in a sentence---- I don't know if the recommendations would have been workable anyway. Probably not. But it just was like, no, you didn't need to do that. Anyway, it's a great place.

MILLER: Thanks so much, David.

BELL: Okay, thank you.

23