Читать книгу Sex, Lies, and Pharmaceuticals - Ray Moynihan - Страница 11

Two 43 per cent

[T]he total prevalence of sexual dysfunction . . . for women . . . 43% . . . an important public health concern.

—Journal of the American Medical Association, 1999

Let us be absolutely clear here. The article in the Journal of the American Medical Association that famously featured the 43 per cent figure¹ gave birth to one of the most pervasive medical myths of our time. The claim flowing from the article—that there is a condition called ‘female sexual dysfunction’ requiring treatment and affecting one in every two women—is an exaggeration as extreme as it is absurd. The evidence assembled in the article never supported such an assertion, as even its lead author attests, yet different versions of this claim have been shouted far and wide ever since. Since its publication in 1999, the influential journal article has been cited by other scientific papers more than 1000 times, and in the wider culture on tens of thousands of occasions.²

‘Bad news in the bedroom, a sex study finds widespread sexual dysfunction’, screeched one popular magazine when the journal article first appeared.³ A management consultancy firm excitedly pointed out that, with 43 per cent of women suffering from this ‘treatable disease’, there must be an awful lot of ‘unmet need’, and the ‘sexual dysfunction market’ would offer great opportunities for companies that sell therapies.⁴ The 43 per cent figure worked its way from scientific journals into medical textbooks and magazine articles all over the world, becoming part of the scientific bedrock for this new disease. Yet no matter how many times the statistic is repeated, it doesn’t make the claim of a medical condition affecting one in every two women any more accurate or any less misleading. The way such figures can come to be so influential is a fascinating part of the story of how modern diseases are made, and FSD is the latest example we have. So where better to start this part of the tale than on a Wednesday in February 1999, the day this 43 per cent figure first emerged into the glare of global publicity.

Ed Laumann’s striking blue eyes shine out at you from his strong sincere face, fringed these days by his shining white hair. The professor of sociology from Chicago University has had a long research interest in human sexuality. He even holds the distinction of having had the US government try to shut down public funding for some of his work. A number of politicians were morally objecting to a large sex survey Laumann had wanted to run, back in the days when conservatives were still in the political ascendency in Washington. His sex survey eventually went ahead with funding from philanthropic foundations, and the results were published in a book in the early 1990s.⁵ Several years later, a selected portion of those survey results would be re-published in detail in the Journal of the American Medical Association, with Edward O. Laumann listed as first author.

The sociologist is particularly proud of the original survey, which was based on a large national sample of interviews with over 3000 Americans. It has, in fact, been highly praised as one of the first to use modern statistical methods of ‘probability sampling’. That’s the method used to try to gather a truly representative group of people in order to give a reliable picture of a whole nation’s sexuality. Alfred Kinsey’s famous interviews with more than 10 000 men and women, while historic, did not employ those same sampling methods, and his findings have been criticised for not being representative.

When the HIV epidemic hit in the 1980s, Kinsey’s work (despite its limitations) was still one of the most comprehensive surveys of American sexual behaviour. Clearly there was a desperate need to update the knowledge and learn more about the sexual behaviour of Americans, hence the motivation for Laumann’s survey. As it turned out, the study found lower rates of homosexuality than Kinsey did; unsurprisingly, it also discovered that a person’s chances of catching a sexually transmitted infection went up dramatically as the number of sexual partners increased.

While the main focus of the national sex study was about trying to better understand and fight HIV/AIDS, it included almost as a sideline some questions about sexual difficulties. Among many other items in the hour-and-a-half-long survey interview, the staff working with Ed Laumann asked women a quick series of simple questions about whether they’d experienced any of seven common difficulties for several months or more during the past year. The questions asked things like whether they’d lacked interest in sex, felt anxious about their sexual performance, had trouble with lubrication, failed to orgasm, came to orgasm too quickly or experienced pain on intercourse. By simply answering yes to just one of these questions, women were then categorised as having a sexual dysfunction. When they were all added up, the number of women in this category reached 43 per cent. And this is the source of the 43 per cent statistic that appeared in the aforementioned article, which has become part of the foundation for a global campaign claiming almost half of all women suffer with a sexual dysfunction that might be treatable with powerful drugs.

The obvious problem is that common difficulties are being confused with medical dysfunctions—a confusion that occurs time and time again with surveys of this condition.⁶ The 43 per cent result came from simply adding up the numbers of women answering yes to a simple question about their experiences, no matter how serious they were. Women were classified as having sexual dysfunction if they answered ‘yes’ to just one of these questions, even if the answer to the other six questions was no. The questions did not ask if they had a mild or severe problem, or whether women had a little or a lot of trouble with lubrication. In the totals from this survey (and in others) researchers are including women with passing problems that might be best seen as healthy adaptive responses to life circumstances. Their findings are estimates of the numbers of people with common experiences; they are not estimates of the numbers of women whose situations are so severe and distressing they may warrant the diagnosis of a medical condition.

Importantly, since at least 1998 the conventional view among those who have been promoting a more medical approach to FSD has been that a woman must be distressed by her situation before she can qualify for a diagnosis.⁷ Yet the original survey questions made no mention of distress. As it turns out, the 1999 journal article actually carried a small but very significant caveat to its findings. It noted, almost in passing, that its results were ‘not equivalent to clinical diagnosis’ of dysfunction. Unfortunately, caveats like this have too often been lost in the gold rush to build this lucrative new condition.

Impressive statistics like 43 per cent can serve very important roles. Being able to throw around enormous estimates of the prevalence of a medical condition directly helps drug companies to claim there is a gaping ‘unmet need’ for their new products. It also helps researchers to attract the attention of funding bodies, and patient advocacy groups to generate interest from journalists keen for a dramatic headline. In short, these figures help bring the recognition each field of medicine so desperately seems to want—and specialists working in this new field of sexual medicine are no exception.⁸

One of the key textbooks about women’s sexual dysfunction actually makes a direct reference to the critical importance of the 43 per cent figure. On page 745 of Women’s Sexual Function and Dysfunction, the editors acknowledge concerns that such large figures may give the appearance that ordinary life experiences are being portrayed as the signs of a medical condition.⁹ Yet, according to the editors of this textbook, the survey that produced the 43 per cent statistic has proved very beneficial in ‘spreading the word’ about sexual dysfunction to those women who qualify for a clinical diagnosis. At the same time though, the inflated nature of the statistic has helped create a strong backlash. Leonore Tiefer, like many others in the field, rejects the claim that 43 per cent of women suffer with a dysfunction, and sees the figure as being extremely unhelpful. But she says with her trademark sense of humour that she finds the 760-page textbook very useful around the office, not least because its size is just right for doing upper body exercises!

Another of those concerned about the size of the 43 per cent figure has been psychologist Dr Sandra Leiblum. She has argued that genuine dysfunction among women is less common than 43 per cent, and that the statistic has contributed to an over-medicalisation of women’s sexual experiences. ‘I think there is dissatisfaction and perhaps disinterest among a lot of women, but that doesn’t mean they have a disease,’ she said in an interview in 2002.¹⁰ Dr Irwin Goldstein, one of the editors of Women’s Sexual Function and Dysfunction, has dismissed suggestions from his colleague that the 43 per cent may indicate the prevalence of difficulties rather than real dysfunction. ‘I love psychologists,’ he told a reporter, ‘but they don’t deal with evidence.’

Back in February 1999, it wasn’t long before Ed Laumann’s pleasure about his latest publication was turning into an experience more akin to pain. Just four days after publication, a major piece appeared in the New York Times. The Times was reporting that the Journal of the American Medical Association had failed to reveal that two authors of the 43 per cent article had some sort of financial relationship with the pharmaceutical giant Pfizer.¹¹ One of those authors was Ed Laumann. As the Times pointed out, the journal article did find that many common sexual problems had social and emotional roots, which drugs couldn’t do much about. But the journal article had also called for the development of ‘appropriate therapies’ for the vast numbers of supposedly dysfunctional women. Those therapies could potentially include counselling and drugs—in particular Pfizer’s Viagra, which had just been approved for men and was at that time actively being studied in women. In other words, Pfizer stood to benefit greatly from the article’s findings of an enormous potential market among women—a great ‘unmet need’. Therefore, any connection between the article’s authors and the company represented the sort of conflict of interest that should have been disclosed, but hadn’t been.

Pfizer had not funded the original national survey, nor had it had anything to do with the later rewriting of results. However, Ed Laumann had joined one of Pfizer’s advisory committees in the lead-up to the launch of Viagra, and that fact was not disclosed in the article. The second author, whose conflicts of interest were similarly not disclosed, was Ray Rosen, the psychologist who’d organised that crucial meeting between drug companies and sex researchers at Cape Cod just a year or so earlier. Two months after the 43 per cent article first appeared, an embarrassing correction was published by the Journal of the American Medical Association. In tiny print, it was disclosed that Laumann had served on Pfizer’s committee, and that Rosen had worked as a consultant or received research support from a total of five drug companies, including Pfizer.¹²

According to Ed Laumann, the idea for the 1999 journal article focusing on the prevalence of sexual dysfunction in men and woman had come from Rosen in the first place. ‘Ray suggested it to me,’ Laumann said. He explained that Rosen used to quote the statistics from the earlier book of survey results, but that a short article in a big journal might be a better way of making people more widely aware of them. Dr Rosen declined to answer questions for this book, so it is unclear how he sees the background to the famous paper or the controversy and criticism surrounding the infamous 43 per cent figure. Ed Laumann, however, was happy to chat about both subjects, over morning tea at a big drug company-funded sex conference in Paris a decade later, in the exhibit hall just across the way from the Pfizer stand.

So what did the sociologist think about the way that 43 per cent statistic had been used to represent an estimate of the numbers of women with a medical dysfunction? ‘That’s clearly a misrepresentation,’ Laumann said with evident frustration. ‘First of all I don’t think that these things are medical dysfunctions, in the sense that they should require active intervention.’ And how did he feel about his figures being used as an indication of the size of the market of women who might need drugs? ‘The problem is you can’t really control how people are going to spin numbers,’ he responded, adding that it would be a mistake to see 43 per cent of women as a potential market for medicines: ‘It’s like dangling a huge carrot out there to say, just look at all those people, half the population, ready to gobble some kind of solution, medication or something like that. It’s a lack of understanding of what it is that really drives those numbers. And what drives the numbers is stress, physical and social stress, exhaustion, not being in a relationship with somebody you care about, so you’re not sexually interested.’¹³

The Chicago sociologist has clearly been stung by the heated controversy surrounding this work and the repeated criticism of the 43 per cent figure. In subsequent articles, he’s been careful to make his caveats more prominent. In a chapter in Women’s Sexual Function and Dysfunction, Ed Laumann and a colleague analysed the results of fifteen surveys designed to ascertain how many women have the condition. They came up with some scathing criticisms of their own about the way in which common difficulties uncovered during these surveys are often portrayed. ‘Despite widespread usage of the term “sexual dysfunction”,’ they wrote, ‘most of these studies rely not on clinical diagnosis of dysfunction, but on self-reports of symptoms or sexual problems.’¹⁴

Laumann went on to emphasise that in almost all the surveys carried out, the high rates of sexual problems apparently being uncovered did not equate to rates of medical dysfunction that might be diagnosed according to the technical definitions. One of the only studies to actually try to find out how many men and women in the population suffered from a diagnosable dysfunction came up with a rate of just 3 per cent. In a sense, Laumann was drawing attention to the inflated estimates of this condition, alerting us all to the myth that has arisen—ironically, partly from one of his own journal articles.

Echoing a wider concern expressed by others, he also criticised the surveys for trying to look at an individual woman’s sexual difficulties in isolation from the context of her relationships. ‘Many studies assessed the prevalence of problems with desire, arousal, orgasm and pain without investigating the nature of the corresponding sexual relationships,’ he wrote in his review. Perhaps most importantly, the sociology professor raised the awkward issue of how to define a sexual problem. He pointed out that what was missing in many studies and surveys was ‘whether the respondent views a particular symptom as a problem’. In other words, leaving aside the inflated estimates of dysfunction, how many of the millions of woman who supposedly have sexual problems—according to these survey findings—are in fact experiencing something they consider to be a genuine problem at all?

This is a question well worth pondering for a moment. When we look behind the claims of high rates of sexual dysfunction— like 43 per cent—it becomes clearer that the studies have actually come up with estimates of the rates of common ‘difficulties’—if they can even be called that. But looking more closely, how many of these common experiences—like a lack of interest in sex, troubles with lubrication, or coming to orgasm too quickly—are even seen as problems by the women experiencing them? And if she does look on them as a problem, does she see them primarily as her own personal problem, or rather something to do with her relationship or her partner? As others have observed, the criteria used by researchers to identify a sexual problem are arbitrary and don’t necessarily relate to the reality of women’s lives.¹⁵

The same year Ed Laumann’s paper was published another sex survey was being set up, organised by a team working in Bloomington, Indiana. This time the survey was being run by the Kinsey Institute, the outfit named after the celebrated sex researcher. Continuing the work Kinsey and his colleagues had conducted decades earlier, the Institute’s general approach is to view the inhibition of women’s sexual response as an adaptive mechanism. Under this approach, a woman’s lack of interest in sex may well be an appropriate and understandable reaction to life circumstances—stress, tiredness or threatening patterns of behaviour from a partner—rather than a sign of some medical malfunction.

Smaller than Laumann’s, this next survey involved almost 1000 American women. It was overseen by Kinsey Institute director Dr John Bancroft,¹⁶ a respected sex researcher and one of the people who had great concerns about this new claim that 43 per cent of women had a sexual dysfunction. Bancroft believed the term ‘dysfunction’ could be highly misleading, making many women feel they had a malfunction when they did not. As to the figure of 43 per cent, Bancroft felt it was outrageous, and did not stand up scientifically.¹⁷ Further, he argued, the statistic was being used wrongly as evidence of a widespread need for medical treatments, including drugs, even though Laumann’s journal article itself had shown that sexual problems were commonly linked to relationship issues and other quality of life concerns. Bancroft was worried that, in the era of Viagra, the complexity of female sexuality might be reduced to narrow questions of how much blood was flowing to a woman’s genitals.

The Kinsey Institute survey asked women a range of questions about their sexuality, their relationships and their physical responses during sex, including questions about lubrication, pain and orgasm. The study found that 24 per cent of women expressed distress about their sexual relationships and/or their own sexuality. Drilling down to what might be causing this widespread dissatisfaction, the survey results suggested the state of a woman’s emotional well-being, and her feelings about her relationship with her partner during sexual activity, were more important determinants of her distress than problems around physical arousal, lubrication or orgasm.

The survey results also indicated that distress about sex did not increase with age, even though interest in sex might decline. This reinforced other evidence suggesting that getting older might bring a slowing of sexual interest or activity, but it also generally means less likelihood of worrying about it.¹⁸ Like others, Bancroft was alarmed that portraying common problems as dysfunctions—as he believed the Journal of the American Medical Association article had done—could make many older women think there was something wrong with them when there wasn’t. ‘While it is good to encourage older couples to maintain and foster their sexual intimacy,’ he wrote, ‘should we be encouraging older women to regard themselves as “dysfunctional” because they have less sexual interest than when they were younger?’¹⁹

Here is a reaffirmation of Kinsey’s celebration of the enormous variation in human sexuality. It is a call to acknowledge the ebbs and flows of sexual life rather than manufacture new medical conditions where the sexual activity of the young becomes the norm for all, and we label those who don’t measure up as abnormal or dysfunctional. Bancroft’s comments are part of a broader concern that the changes in sexuality that tend to happen naturally as we age are being reframed, or re-imagined, as a health or medical problem requiring treatment.²⁰ Yet, at the same time, his perspective readily accepts that some sexual difficulties can be severe and long lasting, and that a medical approach may sometimes be entirely necessary. Criticising the drug companies for trying to portray common sexual problems as treatable disorders doesn’t mean rejecting the obvious benefits of a medical approach and medicines, when needed. For John Bancroft, himself a medical doctor, some malfunctioning in a woman’s sexual response could occur because of the side-effects of a drug, because she was suffering a disease like cancer or because of some failure of her healthy adaptive mechanisms. In such cases, in his view, the word ‘dysfunction’ may well be appropriate, and treatments valuable, including medicines proven to be safe and effective.²¹ Like many surveys, the one conducted by the Kinsey Institute was funded by the pharmaceutical industry, and in this case the company—Eli Lilly—had also hired Ban-croft as a consultant.

The journal article based on the results of that survey was published in mid-2003, and by the end of that year Lilly’s sex drug Cialis would be approved for men. Like Viagra, this drug acts by enhancing blood flow to the genitals—a physiological process the industry hoped would help some of the many millions of women said to have this new dysfunction. Yet as the industry was becoming excited by the profitable possibilities, respected voices like Bancroft and Tiefer were raising questions about whether there really was a mass market of women with a medical condition requiring drug treatment. If distressing sexual difficulties were largely caused by relationship and emotional factors, rather than a medical dysfunction of poor blood flow, where on earth was this market for drugs?

John Bancroft was asking two very pointed questions: When might a dysfunction better be called a difficulty? And when was a difficulty better seen as normal life anyway? His journal article analysed the ways in which several other influential surveys had dealt with these awkward questions. He concluded that many of the women supposedly found to have sexual problems through sex surveys often didn’t even regard themselves as having problems or weren’t distressed by them. In other words, surveys were not only confusing difficulties with dysfunctions; they might have even been finding problems where none existed. Which begs another awkward question: to what extent are these surveys, which often generate a lot of coverage in the media, actually helping to paint a picture of a new epidemic of sexual difficulties that may not in fact be there in the first place?

Bancroft went so far as to suggest that until there was an agreed way to accurately define whether a woman had a problem with her sexuality or not, running big surveys to assess how widespread they were was a ‘hazardous’ activity. He also called for a whole new approach to assessing sexual problems, in which a woman’s own description of her experience is seen as more important, ‘rather than her answers to questions based on preconceived concepts of female dysfunction’.

Just a few short months later, that call for a new approach to assessing sexual concerns was heard and heeded on the other side of the Atlantic Ocean. A British team based in London decided to take the radical step of conducting a sex survey that included questions asking women straight up whether they thought they had a sexual problem.²² The plan was to try to find out how many women might end up with a label of dysfunction if their doctors used a standard approach, and compare that with the number of women who actually saw themselves as having a sexual problem. The researchers recruited 400 women from several general practices across London. The research project was unusual, and its methods rather complicated, but they went something like this.

In order to see how many women might be diagnosed by their doctor under a medical approach, the researchers used a questionnaire and a definition of sexual dysfunction derived from an international manual of diseases used in Britain— similar to the American DSM. (A key difference between the British and US definitions is whether or not the woman must experience distress. In the British definition, this requirement is absent.) Unlike other studies of this kind, the British researchers also asked the women directly whether they thought they had any sexual problem, even of a mild nature, and how distressing that problem might be. In addition, the women were asked whether they considered their problem to be a medical one— like arthritis—an emotional one or relationship related—like a lack of time together with a partner. The findings were stunning.

When the researchers looked at the results from the questionnaire used to assess the women, they found that a whopping 38 per cent might qualify for a doctor’s diagnosis of sexual dysfunction. But that figure then fell to 18 per cent if the woman herself was required to think she had a problem. When it was also required that the woman saw her problem as distressing, the figure fell again—this time to just 6 per cent, a world away from 38 per cent.

In other words, only a tiny proportion of women who were ‘diagnosed’ with a sexual dysfunction during this study saw themselves as having a sexual problem that distressed them. This is clearly only one small study with an unusual scientific approach, but it raises fascinating questions about the value of the definitions of these conditions. If only a very small number of women labelled as dysfunctional actually see themselves as having distressing sexual problems, what does that say about the definitions of the dysfunction? A major reality check is surely looming.

As to what the women saw as the origin of their sexual problems, the innovative London study found relationship and emotional issues were the most commonly perceived causes. For almost two-thirds of the women in the study who said they had a problem, difficulty with their relationship was seen as one of the factors. This doesn’t mean those difficulties aren’t genuine, but it does raise doubts about the relevance of drugs in treating them. Echoing the approaches of John Bancroft and Leonore Tiefer, the British researchers concluded that ‘reduced sexual interest or response in women often appears to be an adaptation to stress or an unsatisfactory relationship.

Many such adaptations may be short term and not classifiable as dysfunction.’ This is a point of view shared by a lot of sex researchers who are uneasy about using the word ‘dysfunction’ to describe common problems.²³ ‘Most women regard their sexual problems as something to do with relationships,’ said Dr Michael King, the professor based at University College who oversaw the London survey and a psychiatrist with an interest in psychosexual medicine who regularly sees both male and female patients. ‘I find very disturbing the drive to find a pill to medicalise everything to do with sex.’²⁴

Unlike other studies of women’s sexual problems, this British research was entirely free of pharmaceutical industry funding, and none of the researchers had any connections to any company. Michael King is one of the small but most likely growing numbers of doctors around the world who has decided not to accept any more free lunches from companies trying to woo his favour. He avoids financial relationships with them, and he’s banned them from sponsoring meals in his medical school department. ‘I’m not anti-pharmaceutical companies,’ he says. ‘They are incredibly important. I just don’t like researchers getting so close to them.’ While he fully recognises the need for doctors to work with drug companies testing drugs, he’s worried when they join forces to conduct surveys or develop diagnostic tools, because of the potential that the marketing and the science will get mixed up. On a positive note, he argues that with all the attention drugs like Viagra are producing, people are more likely to be open with their doctors, even though they don’t necessarily see their sexual problems as medical ones. Importantly, he points out that ‘they are still afraid to talk to each other’. He is hinting that a frank conversation with one’s sexual partner may often help a lot.

While Michael King might choose to avoid pharmaceutical companies’ free lunches and financial ties, other researchers, however, very happily continue to work with industry on surveys of this condition. And sometimes the drug companies don’t just fund the studies: they actively initiate them, run them and help write up their results. When sociologist Ed Laumann published the results of a later survey of sexual problems, this time around it was global—and this time it was directly funded by Pfizer.²⁵

Funding wasn’t where the company’s role ended, either. One of the people who helped analyse and interpret the results was actually a Pfizer employee. And as Laumann explained it, the motivation for the huge international survey actually came from within Pfizer, which orchestrated it and paid for it. ‘It was a marketing effort,’ said Laumann bluntly, to help the drug company ‘gauge the scale of interest’ across the markets of many different nations.²⁶

That study surveyed close to 30 000 men and women in almost 30 countries. It used similar questions to the ones Laumann had used in his original American survey, although this time round the results were not presented as dysfunctions, but rather as sexual problems. The rates were again extremely high, with findings that in some parts of the world over 40 per cent of women had experienced some problems. As an example, according to the results from countries in Northern Europe, almost 20 per cent of women apparently had the problem of failing to reach orgasm, with that figure rising to over 40 per cent in Southeast Asia. Those statistics sound horrifyingly high, and they give the sense the world is living through a plague of sexual problems. But it is worth taking a closer look at how those figures were calculated.