The Psychophysiology of Sex

Discussion Paper

Discussion Paper julia r. heiman This morning focuses on clinical populations and clinical problems. Several issues are highlighted. One is measurement, a continuing theme of this meeting, which increases in importance when describing clinical populations and treatment effects. A second issue involves the de¤nition of change. It is no longer adequate to have simply a signi¤cant change statistically; a clinically meaningful change is increasingly the desired endpoint. A third issue involves the sample selection. How that gets de¤ned varies. For those of us who conduct pharmacological trials, the degree to which one works with not only a self-selected sample but also a highly screened sample deserves comment. Recent trials in investigating sildena¤l in women excluded 80% of screened women, even though they had the diagnosis in question. The consequences of this intensity of screening are worth discussing. A fourth issue is the current social environment; using sexually explicit materials is socially sensitive, as is the testing of experimental drugs. For example , on the latter point, revelations about the safety problems of pharmaceutical products not only have important human consequences but also impact potential subjects’ willingness to volunteer for studies involving experimental agents. Let me ¤rst talk about Ray Rosen’s very thorough summary of outcome assessment, particularly in the area of psychopharmacology. Ray started out with an example, actually an example from 1994, which was a study by Munoz, Bancroft, and Beard. This is a well-designed study on men. That study had a complicated procedure. They included a number of measures of penile response. I think we need to get beyond simply measuring a mean and/or maximum response, particularly for clinical ef¤caciousness . Can we measure latency, and how con¤dent would we be to measure latency to a certain criterion response? What is interesting about that Munoz et al. study is that subjective measures were seen as secondary outcomes. Penile measures were seen as primary outcomes. That is true, I have the impression, for most of the research in men. The primary measure has been the genital outcome, the secondary measure has been subjective. Ray’s paper also mentions the Goldstein et al. (2000) article where erectile rigidity was measured by a buckling procedure. What happened to that measure? Was it not useful or just too cumbersome? When have you 425 seen a sildena¤l or vardena¤l study using the buckling force procedure to measure outcome? It was heavily used for a while. We do not hear of its relative value for clinical studies. In general, what I felt Ray’s paper is really capturing, and in some detail , is the issue of what would be desirable in study design and methods used for testing men. What you see, I think, across time, is that less money is being spent on these studies, so there is a contraction of the number of measures used. Now that may be good, but it may not necessarily further the science. In studies with women, sexually dysfunctional women are increasingly being studied, whereas 3 years ago, we had the greatest amount of data on healthy young volunteers. So, we’re a bit more weighted on the healthy side than on women with sexual dysfunction. As for women with sexual dysfunction, the focus is almost exclusively on women with sexual arousal disorder, in large part due to the success of sildena¤l in men. Now it turns out that female sexual arousal disorder is very dif¤cult to de¤ne. It’s de¤ned in the DSM, but it clearly overlaps with desire disorders. How often is genital arousal disorder present without subjective arousal problems , which is what the DSM implies in the focus on genital symptoms? This distinction may have substantial consequences for how we measure arousal and change in arousal in the lab. In addition, hormonal status has become increasingly important in the study of women. So, for example, in the sildena¤l studies that have been done, it has become clear that inadequately androgenized and inadequately estrogenized women do not respond to sildena¤l. One wonders how many other types of compounds are being developed for which this might be important. So I would encourage the inclusion of repeated measurement of hormonal status, whether women are on hormonal treatments or not. I can’t resist talking about self-reported arousal–vaginal response differences . What women say about how aroused they are is obviously important . If we only have a vaginal change in a...