C- Transcript of Interview with an End User (#0452) INTERVIEWER: I'm going to be asking you a series of questions about searches that you've done, one search at a time. So if you could think of a recent instance in which the information you obtained through a MEDLINE search you conducted was especially helpful in your work. Can you think of a specific search? RESPONDENT: Um, as you get older you can't remember. I think I did one on strep throat. INTERVIEWER: OK. RESPONDENT: Streptococcal Group A beta-hemoglobin infection. INTERVIEWER: What was the situation that led you to do this search? RESPONDENT: Um, I'm the maternal and child health coordinator and the head of pediatrics here at the [Name] Hospital in [City]. And I had to revise the old policy and procedure for managing strep infection. Is that confusing? INTERVIEWER: No, no, not at all. So, you were revising your procedure for strep... RESPONDENT: Yes. Procedure for the clinical management of strep throats. INTERVIEWER: And what was the old policy that needed changing? RESPONDENT: Um, the policy isn't--hasn't changed much, actually, but I just wanted to make the medical staff and the nursing staff aware of some of the newer findings, the newer information about strep throat. INTERVIEWER: And when you went to do the search, what was the specific information you were searching, the question you had in your mind? RESPONDENT: There were a number of them. One of them was the proper antibiotic treatment, the second thing was whether or not non-Group A strep infection had to be treated as vigorously, the third thing was whether we should treat promptly or wait till the culture comes back--do you want me to list all of them? INTERVIEWER: Sure, I'm getting them. RESPONDENT: Another one, the fourth one, would be what you do with the carrier state, or those who are repeatedly positive. I think those were the main issues that we wanted to clarify. INTERVIEWER: And you were interested in the Group A strep? RESPONDENT: Yeah. Called GAS--Group A Beta-Hemoglobin Strep Infection. INTERVIEWER: And how does that differ from the non-Group A? RESPONDENT: Well, the non-Group A are usually things like Group C, which can cause family outbreaks, but don't cause rheumatic fever. Group A strep causes rheumatic fever if it's untreated, which the other groups don't. INTERVIEWER: OK, so the other one is less possibly dangerous. RESPONDENT: Yes, that would be correct. INTERVIEWER: OK. And why did you choose to do a MEDLINE search instead of consulting some other source like textbooks or journals or colleagues? RESPONDENT: Well, I did consult--well, I'm the expert here, so there's no colleagues--actually I did, I consulted one of the internists who is an infectious disease person. The textbooks are four years out of date, I did consult textbooks. I did consult the Red Book, which is only a year old, which is a good source. But I wanted to read some of the literature myself and I find that if I go to the nearby university, there's a lag time of, you know, a week or so, and then they charge us much more money than it would cost if I did the search myself. INTERVIEWER: OK. And how did you go about carrying out the search? You said you used Grateful Med, and then did you use MeSH headings or textwords? RESPONDENT: I used textwords. I'm in the process of getting the new revised brochure about how to use MEDLINE so I can use the MeSH better, but I just used the text and I ended up fumbling around, you know, I have to do a couple of searches because I don't do it right. In other words, you know, I put it in, I used regular text, and then it's too broad a category, too narrow a category, and then I gotta redo it--but I'm learning. INTERVIEWER: And do you remember what the textwords were that you used? RESPONDENT: I think I finally came down to searching under "streptococcal diseases." INTERVIEWER: And did you use "streptococcal disease" AND anything OR anything? RESPONDENT: I don't think so. I think what I did--I had trouble, if I used Group A beta hemolytic strep it didn't work, I seem to have a little trouble, and I probably wasn't searching correctly. INTERVIEWER: OK. So you found the best way to do it with just "streptococcal disease" alone. RESPONDENT: Correct. INTERVIEWER: Did you search backfiles? RESPONDENT: No. INTERVIEWER: And did you-- RESPONDENT: I would have searched backfiles except that--and I've spoken to the people who made Grateful Med before--is that the old version, Version 3, I think, wouldn't run on my IBM--not clone, but my IBM computer--and a whole bunch of other people I know IBM computer--I wrote in and they gave me some kind of a quick fix but it never seemed to work, so I ended up using the old Version 2 or something where to do backfiles it was difficult 'cause it didn't have the right program in it, so with the new Version 4, I should be able to do more backfile searching. INTERVIEWER: Mm-hmm. And did you use any kind of limiters, like "English only" or "reviews only"? RESPONDENT: Yes. I used--I first used--I always use "English only" but--because those are the ones I'm more interested in, and it's easier, because I can't translate 'em or anything. But I think I did "reviews only" then I did one without a review. I did the "reviews only" first and then I did another one without the reviews. INTERVIEWER: OK. And what was the information that you obtained as a result of the search? RESPONDENT: Well, it answered all my questions! I know it sounds like a testimonial, but it did. You know, I found out about Group C, I found out about chronic carriers and this and that--there were a whole bunch of real good articles. Some of which I actually had, and it reminded me to go look in my own files for copies. INTERVIEWER: And can you remember any of the specific information, like what did it say about proper antibiotic treatment... RESPONDENT: Well, sure, I mean--it said basically--well, lemme just give you the things that I found that we were doing that wasn't right. INTERVIEWER: OK. That would be great. RESPONDENT: Number one, as we were using Bicillin CR, which is a long and short acting combination, when we probably should have been using almost all Bicillin LA and only Bicillin CR if the kid's under 12. Number two, the treatment of the carrier state for recurrences using Cephalosporin and also Rifampin was relatively new knowledge. The fact that you cannot predict the presence or absence of Group A strep infections from the number of colonies--we used to think that if you had a few colonies it was a carrier, if there was a lot of colonies you had an active disease--you can't tell. And, let me see what else--the other thing was the Group C thing, that being the cause of rheumatic fever, but just as it being treatable, in the sense that you can get a strep throat, but you won't get rheumatic fever from it. There were a few other minor things, but those were the bigger things, I think, that came out of it. INTERVIEWER: In what specific ways was the information helpful in your decision-making. What did you do with the results of what you got? RESPONDENT: Well--I rewrote the policy. You know, I wrote a--what I did now, is I wrote a, like a fact sheet that'll give the kind of a like a summary of recent and old clinical findings that are significant for people to know, indications for throat culture and treatment. See, one of the reasons why we're doing this is we thought we were taking too many throat cultures. And it turns out we're probably not taking too many throat cultures. And that I found out too, but I knew that before, actually. Um, that you have to have a certain percentage of negatives--you can't, in other words if you have--if you do throat cultures on school kids like we do here, and you have 90% of them positive, you aren't taking enough throat cultures. You're being too selective. You need to have a probably like a 25 or 30 percent, you know, positive, and you have to have like 75 percent negative, otherwise you're not--you're being too selective. INTERVIEWER: So that anything that remotely looks like it might be strep you culture. RESPONDENT: Well, I wouldn't say remotely, but we establish criteria for when you would culture a throat, and I wrote it down, you know, in a very simple way, so that the nurses and the doctors and non-pediatricians and the nurse practitioners and everyone can follow it. So we established guidelines that were medically correct but simple. We didn't go into these long-winded algorithms, I made it very simple so that anyone can pick it up and just say "Aha! Sudden onset of a fever of 102, bad sore throat, pus on the tonsils, swollen glands, exposed to strep, past history of rheumatic fever--treat," rather than get into a long-winded philosophical discussion, which clinically you can't function that way. People just don't make those decisions. INTERVIEWER: Mm-hmm. So it was like sudden onset of fever, presence of pus on the tonsils, red throat-- RESPONDENT: --swollen glands, what's called an enanthem, or petechiae on the soft palate, recent exposure to strep, past history of rheumatic fever, would be reasons to treat. INTERVIEWER: OK. And-- RESPONDENT: You could take a culture I mean and treat, and not wait for the culture to come back. INTERVIEWER: OK. RESPONDENT: Or "treat before culture is returned." INTERVIEWER: And what was the impact on the situation of having the information? RESPONDENT: I can't tell you that, because I have to give the inservice, and we have--I mean, this, a little, we just finished it like a week ago. I feel good about the policy because it's not much different than the old policy, to be honest with you. The old policy was also very practical, and the old policy has stood the test of time, in the sense that not much has come up to change. We've changed some of the treatments and the understanding of it from doing a search like this but the management of it really hasn't changed a heck of a lot. It's the same thing that I established four years ago. INTERVIEWER: And what do you think might have happened if you didn't do the MEDLINE search? RESPONDENT: Well, we would have fumbled around again with the same old policy, and people would have been acting on misinformation and lack of information. I think this way it's clarified that I feel as the head of maternal and child health policies and procedures here, I feel comfortable in that this is a reasonably good policy. And as long as I feel comfortable, then I can tell people to carry it out. INTERVIEWER: And what do you think the outcome of this situation will be? What do you think will happen to the patients? RESPONDENT: Well, I think they're going to get better care, although they got good care before, but I think the care will be improved in the sense that all of us will have a better idea, there'll be a consistent approach instead of one provider saying this and one provider saying that, I think the nurses will feel more comfortable knowing that there's a consistent policy, they don't have to go up to three doctors to find out what to do, and I think if we are--which we're not planning on, but we are going to do any kind of a clinical review or a QA thing--quality assurance, of how we're treating sore throats, at least we've got the beginnings of 1) a policy and procedure, and 2) we can follow these people. So it kinda tidies up the whole thing. It's a common illness we see in the outpatient department, plus we have it in school, the kids nearby. INTERVIEWER: Especially at this time of year. RESPONDENT: Yeah. INTERVIEWER: Uh, how long ago did you do the search? RESPONDENT: I can't recall exactly, I'd say 3 weeks ago. INTERVIEWER: So it's certainly within the last month. RESPONDENT: Oh, yeah, it's in the past month. INTERVIEWER: And you'd say this would be for a type of patient? or would you call it for administration? RESPONDENT: It would be both. It was an administrative need, to establish the policy and procedure, and also is for specific patients' illness. INTERVIEWER: Right. And where was the search performed? RESPONDENT: You mean in [City], [State]? INTERVIEWER: No, no, did you do it in your home or your office? RESPONDENT: I did it in the office. INTERVIEWER: OK. When you obtained the citations, did you request the abstracts online? RESPONDENT: Yes. INTERVIEWER: OK. And did you then go and look at the articles themselves? RESPONDENT: Umm, in the articles that I had, I was surprised that I could find actually a fair number of them here, yes. INTERVIEWER: OK. And you used Grateful Med the form screens? RESPONDENT: Form screens... INTERVIEWER: As opposed to using command language. RESPONDENT: No, I used the forms screens. INTERVIEWER: OK. And do you know what version of Grateful Med you used? RESPONDENT: Uh, I think it was 2...It actually may have been the latest one, 4, I'm not sure whether I had put that one on yet. It was either 2 or 4. Actually, it was 4. INTERVIEWER: OK. And have you always accessed MEDLINE this way? Using Grateful Med ? RESPONDENT: No. In the beginning, years ago, I used Knowledge Index. INTERVIEWER: What is Knowledge Index? RESPONDENT: It's a DIALOG-- INTERVIEWER: Oh, it's using DIALOG. RESPONDENT: It's a DIALOG, and then you get into using Knowledge Index which is a version of it made for--it's easier to use. But I haven't used that--since Grateful Med came along, that's all I've been using. INTERVIEWER: And it sounds like you have done mediated searches where you've gone to the library and had them search for you. RESPONDENT: Yes. * * * [Interview continued with second incident report] INTERVIEWER: OK. Can you think of another situation in which you found MEDLINE either very helpful for your work or alternatively where you had a lot of problems and you found it was not helpful? RESPONDENT: Uh, well, someone asked me to look up, I think, some blood disorder. INTERVIEWER: Was this effective or ineffective? RESPONDENT: Well, it turned out to be ineffective, and I think that the reason was again, that I didn't sit down and--she came into the office and wanted it done quickly. Usually I sit down and go over the thing, and I--I forget the word you folks used, but I kinda designed the search, kinda did it quickly and it I got like no responses or 5,000. It was either un--it was either zero or overloaded. I didn't design the search properly. INTERVIEWER: What was the situation that led to the search--somebody on the staff asked you for some information on blood? RESPONDENT: She came up and says, We have a patient with this thing, can you look it up for me? INTERVIEWER: Do you remember what the thing was? RESPONDENT: It was thrombocytopenia, I think. And I don't know why--it's a relatively simple search, I don't know why I couldn't --I think I must have in a hurry typed it in wrong. INTERVIEWER: So the patient had thrombocytopenia. And why did they need to get--what was it -- RESPONDENT: I think they just wanted to get some of the latest literature--'cause the textbooks are usually behind the times. I mean that's the main--that's why I like this-- INTERVIEWER: And do you know what the question was that the person was asking? RESPONDENT: No. It was just that she wanted to get the latest information on it. INTERVIEWER: What would be the symptoms of thrombocytopenia or what would-- RESPONDENT: Bleeding tendency. INTERVIEWER: OK, so they present a bleeding tendency and the doctor wanted the latest information on what--diagnosis, treatment-- RESPONDENT: Correct. INTERVIEWER: And the textbooks were out of date. RESPONDENT: ...I'm going to have to beg off, because I have to make rounds. INTERVIEWER: OK, let me just ask you a quick series of questions and then we'll be done. How long have you been searching MEDLINE yourself? RESPONDENT: Whenever Version 2 came out, I would say 3 years. INTERVIEWER: And how many MEDLINE searches do you do yourself--have you done yourself? 11-20, 21-50-- RESPONDENT: I'd say 11-20. INTERVIEWER: OK. And how many do you do in a typical month? RESPONDENT: Oh, I probably do one every other month. INTERVIEWER: And you were self-taught, or did you take a class? RESPONDENT: Self-taught, like everything else. INTERVIEWER: And you have an MD, or an MD/PhD? RESPONDENT: MD. INTERVIEWER: In what specialty? RESPONDENT: P Ed. INTERVIEWER: And what percent of your time do you spend in patient care, teaching, research, administration-- RESPONDENT: 100%. INTERVIEWER: 100% in patient care? RESPONDENT: Oh, you want to know how much in each one? INTERVIEWER: Yeah. RESPONDENT: Oh, I probably spend 80% in patient care and 20% in running programs. I work for [Government Agency]--we do public health, so there's a fair amount of administration. I don't mean that I supervise anyone, or--I supervise some people--but I end up designing policies and procedures and--meetings--and stuff like that. INTERVIEWER: And so you say you're 100% with a government agency. RESPONDENT: Yes ma'am. INTERVIEWER: And how large a community does your practice hospital serve? RESPONDENT: Uh, about 20,000 [Population Group]. INTERVIEWER: OK. Well, thank you very much for your help. Bye-bye. C- Transcript of Interview with a Mediated User (#U024) INTERVIEWER: We are interested in recent MEDLINE searches that were especially helpful in your work or ones that were unsatisfactory. And we are also especially interested in searches that have had an impact on patient care. The way the interview is structured I am going to ask you a series of questions about searches that you have requested. I'd like you to describe them one at a time and the questions will relate to the circumstances that led you to do the search, how it was conducted (as best you know), the information you got, and how you are using the information. Can you think of a recent time when the information you got through MEDLINE search was especially helpful with your work? RESPONDENT: OK, I'm just now making some notes down. I've done a few--uh--I'll try to remember them. INTERVIEWER: Are these ones you did yourself or did you request them? RESPONDENT: These are ones I requested of our librarian to do them. They didn't give you a list, did they? I have got three that I remember; I know I've done more than that. One was actually quite a while back--it was on Crohn's disease and nutrition and so on, and that was directly related to a patient. INTERVIEWER: OK. What situation led to you this? RESPONDENT: A patient who was very ill with Crohn's disease and had basically lost some weight and was on a very dangerous trail. INTERVIEWER: And so what problem arose; why did you do the search? RESPONDENT: Partly because I wasn't familiar with all the latest ways of managing the problem and the specific ways of dealing with the nutritional complication of Crohn's. INTERVIEWER: So when you say the problem, you mean her weight loss? RESPONDENT: She also had some various intestinal problems as well as depression. INTERVIEWER: So you were interested in the search of all these various aspects--weight loss, depression, and the intestinal. Why did you chose to do a MEDLINE search in this case instead of consulting some other information source such as textbooks, colleagues, or your journal collection? RESPONDENT: Mainly to try to find the latest information and also good review articles. I did look in the textbooks, too, but it doesn't really have the depth of information. INTERVIEWER: OK. Were colleagues of any help? RESPONDENT: Not a huge amount. Not with the kind of... I guess partly, too, I liked to get a base, I mean the general recommendation, but not the specific research that backs it up or the things that you can look back on later. I felt like I wanted that. INTERVIEWER: OK. How did you explain your information? RESPONDENT: I just got a call; either I could call you back or give me a call in about five minutes or so. INTERVIEWER: Sure, I'll try back in five minutes. INTERVIEWER: [Interview resumed] That search when you asked for it, how did you explain what you wanted to the librarian? RESPONDENT: Oh, this again is going back a little. I think what I said was something about Crohn's and complication of nutrition, something like a conjunction of some of those. INTERVIEWER: OK. Do you know how the search was carried out? Do you know much about that? RESPONDENT: No, they just go ahead and do it. We get the printout which they used, I never look at that. INTERVIEWER: Do you know how many years back they looked? RESPONDENT: I'm not positive. I think in general they may have looked either three years or five years. Although I am not sure. INTERVIEWER: And what information did you get from the search? RESPONDENT: Basically, there were several good review articles and then I got some specific ones about nutritional work and I think one was about intestinal complications. INTERVIEWER: In what specific ways was the information useful to you? RESPONDENT: I see, I guess the main thing was the background information and thinking about ... interpretation of background and thinking about my patient. The specific things we used with her -- if this one wasn't so far back -- a lot of ... forgotten ... an individual patient ... a kind of general issue. But this particular one I know we went ahead and did Hyperal for a while, and I actually can't remember now. INTERVIEWER: What does that mean? RESPONDENT: It means having a central venous line into the chest that you can give sugar and fat solutions and proteins. INTERVIEWER: So you don't know or remember if that was based on recommendations on the articles or-- RESPONDENT: No, I can't. I know that they discussed it some, but I really can't recall how much I got from the articles and how much I used in deciding... INTERVIEWER: OK. How do you spell Hyperal? RESPONDENT: H-Y-P-E-R-A-L. INTERVIEWER: And what was the outcome for this patient? RESPONDENT: It was good. She went home, gained weight, and actually got pregnant. INTERVIEWER: And what is the prognosis for Crohn's disease? RESPONDENT: With Crohn's it varies. There are some people who will kind of remit and they will not have much more problem and there are others who get very bad and eventually need lots of surgery and they die from it, and there are others who just kind of waxing and waning. INTERVIEWER: Do you know what it was for this patient? RESPONDENT: This patient still has active disease and she will probably eventually need surgery, but we did not have to do one. We felt we might have to at the time and it has now been over a year that we have avoided having to do it. INTERVIEWER: OK. How long ago did you do this? When was this search done? RESPONDENT: This search was--I believe it was early last year. INTERVIEWER: And, was this your own patient? RESPONDENT: Yes. INTERVIEWER: What service? Was it inpatient? RESPONDENT: It was medicine. INTERVIEWER: Where was the search performed? RESPONDENT: [Initials of medical center] and ... INTERVIEWER: Who performed the search -- was it a medical librarian? RESPONDENT: Yes. INTERVIEWER: Did you ask for any abstracts when you got citations? RESPONDENT: I didn't ask for the abstracts, I asked for actual articles, not citations. I got a list(?) of very brief titles and then I kind of chose from that. INTERVIEWER: Do you know what system they used at ...? RESPONDENT: No. INTERVIEWER: Did you talk to the searcher, did you fill out a search request form, how did you-- RESPONDENT: What I usually do is talk and I write something down, it's not actually a question. I put something down. INTERVIEWER: And were you present when the search was performed? RESPONDENT: No. INTERVIEWER: OK. I now have a few general questions about your experience with MEDLINE and your practice. How long have you been requesting MEDLINE searches? RESPONDENT: When I was in medical school, I think I did one. But basically since I got to ... probably the second year in a row since I've been doing it--so that makes it a year-and-a-half now. INTERVIEWER: Could you estimate how many MEDLINE searches you have requested total? RESPONDENT: I think it's been about six or seven. I've currently remembered four that I have written down, but I think there were more. INTERVIEWER: OK. How many searches do you request in a typical month these days? RESPONDENT: I probably request only one every two to three months. INTERVIEWER: What's the highest educational degree you have received? RESPONDENT: M.D. INTERVIEWER: And your specialty? RESPONDENT: Family practice. INTERVIEWER: And what percent of your time is spent in the following professional activities? I have patient care, teaching, reseach, and other. RESPONDENT: Patient care is probably 90% and teaching 10%. INTERVIEWER: And what percent of your time is spent in the following work settings? We have self-employed, group practice, HMO, medical school, hospital, and other. RESPONDENT: I suppose it would be -- I don't know how you would say residency in there. Let's see, I guess 60 to 70% of my time in the hospital, 30 to 40% of my time in the clinic. INTERVIEWER: OK. And do you work in [city] and [city]? RESPONDENT: I'm down in [city] this month. We have some required rotations down here. INTERVIEWER: What is the population size of [city]? RESPONDENT: Population is about 50,000 or 60,000. INTERVIEWER: About 50,000? OK. Any other searches that you can think of that have been especially helpful with patient care? RESPONDENT: I did one search about VBAC, which is a vaginal birth after Caesarean section. INTERVIEWER: How do you spell that? RESPONDENT: V-B-A-C. Vaginal birth after having a Caesarean section. INTERVIEWER: What was the situation that led you to do the search? RESPONDENT: That was mainly the general practice situation in [city] at our hospital. Which is that in general they discourage doing that and that they want kind of once someone has had a C-section, go ahead and have another C-section. And I had heard or read some little bits and pieces about how things were going on the popular news that actually were saying that it may be better in terms of morbidity and mortality to do a vaginal--to have a trial vaginal birth after C-section. So I wanted to find out more of the actual research and information on the issue. INTERVIEWER: So you said at the [city] hospital they--the policy generally was to follow a C-section with another C-section. OK. Just thought I would check. Again, why did you choose to do a MEDLINE search in this case? RESPONDENT: Because I didn't have--the issue is new enough that it is not in the textbooks. I looked in Williams and it wasn't there. And also because, in specific terms, in [city], which you know is not a university-type place that generally may have all the latest specialists that are up on it, on the absolute latest things, and because the specific policy there is to go ahead and C-section, I wanted to find out what the general medical community's feeling was on the issue. INTERVIEWER: OK. How did you explain your information need to the searcher? RESPONDENT: I believe it was specifically VBACs and I probably used C-section. I don't think that one needed too much else. I can't remember if I wanted specifically about safety or mortality because most of the things talk specifically about that but I can't remember if I put those in. INTERVIEWER: Again, do you know how the search was carried out? RESPONDENT: No. ... I get the information and the printout. INTERVIEWER: What information did you gain from this search? RESPONDENT: I got some specifics on studies that indeed showed that the morbidity and mortality was less on VBAC than on routine C-section and also had got, along with the search specifics that I was looking for, there were a few other different articles that I don't remember the details ... INTERVIEWER: In what specific ways was the information helpful to you? RESPONDENT: It (was) mostly my own knowledge and what I am going to want to do and can't, and it doesn't really affect and not really able to affect what I do with my patients at this hospital right now. INTERVIEWER: OK. So it won't immediately affect your (patients). RESPONDENT: I'm afraid not immediately because of -- I mean, what I can do, what I was able to do was was explain to my patients what the different issues were, but that if they really wanted to have a vaginal birth, basically at this hospital they would have to refuse permission to have a Caesarean section. INTERVIEWER: OK. So if the patient requests the vaginal, they could go ahead. RESPONDENT: Basically, what they have to do when somebody comes in and tells them they should have a Caesarean section because there's a danger not to, then I can tell them--all I can do is give them the information so that if I'm not there, which I can't always be there, I don't want that ... INTERVIEWER: Has this affected any patients so far? RESPONDENT: There have been -- I've had about three or four patients who had initially desired a vaginal delivery and discussed it with them. One of them ended up having a complication where they needed a Caesarean section. One of them I wasn't there and she kind of got threatened into having a Caesarean section, and another one ended up deciding that she preferred having the Caesarean section anyway. It's more been that I was able to talk with them and to feel that I was able to give them the right kind of counseling. INTERVIEWER: How long ago did this search happen? RESPONDENT: This one was around May or June of last year. INTERVIEWER: And would you describe these as your own patients? RESPONDENT: Yes. INTERVIEWER: And, on what service were you seeing them? RESPONDENT: This was through my clinic. INTERVIEWER: Where again was this search performed? RESPONDENT: [City]. INTERVIEWER: And did you ask for any abstracts with this search? RESPONDENT: No. It was again from the actual articles. INTERVIEWER: Did you speak with the searcher or did you do the same thing where you filled out a form and wrote down some notes? RESPONDENT: Basically we wrote it down. INTERVIEWER: Any other searches that you can think of that have been especially helpful or ones that were unsatisfactory? RESPONDENT: The only ones that have been somewhat unsatisfactory have been more that I haven't been able to get all the articles by the time I needed them. For example, I was giving a talk on sexism in medicine and did a search about two weeks ahead of time and the librarian was only able to get two or three of the articles that I was interested in by the time of the talk. But that's basically all the actual (I have to think now)... Actually, on that particular one too, it's hard to know how many articles have been written on the subject, but I was able to get as much--there wasn't as much on the search list that I would have been interested in looking at that would have helped with the talk as I would have liked. I don't know if ... nothing has been written. I wasn't able to think of the right code word. INTERVIEWER: Would you say that it was an unsatisfactory search overall, though? RESPONDENT: I would probably more say to you that I got a few good articles out of it. [Only two incident reports per interview are included.]