by David Healy
DXA scans offered a golden opportunity to redefine osteoporosis, just as weighing scales did with refashioning standards of beauty. Women now were said to have the disease if their scans showed that bones in some part of their body had densities that were a standard amount less than the densities found in women in their twenties. All of a sudden one-third of postmenopausal women found themselves diseased.23 A further large group of women—and their doctors—were faced with another problem. These were the women who fell in between the supposed optimal bone state of women in their twenties and the new diseased state. For women whose bones fell between these states a completely new condition was invented, osteopenia (literally, “less bone”). Women with osteopenia are at no increased risk of a fracture compared to their “normal” peers.
Two factors combined to ensure that the transformation in our perceptions to make room for this disorder would be particularly rapid.
One was that two companies, Merck and Procter and Gamble, were competing to be the market leader with products from a new drug group—the biphosphonates. Merck's Fosamax was up against Procter and Gamble's Actonel in a billion dollar battle. Both companies saw DXA scans as the way forward to increased sales and they competed to provide scanners for free to doctors—who of course could charge a fee for scanning.24
The biphosphonate drugs conveniently provided an answer to the image of bone thinning—offering, at least in popular understanding, to remineralize bones. Given to women with severe reductions in bone densities, the biphosphonates can reduce what are called fragility fractures—hairline fractures primarily of the vertebrae that are picked up on X-ray without the person ever being aware they have had anything wrong. However, there is no difference between those on the drug and those not in the number of women who present to their doctors with an obvious fracture. Somewhere between 80 and 90 percent of women to whom biphosphonates are now given are unlikely to get even the X-ray changes in the rate of hairline fractures, and in some of those with minimal reductions in bone densities the biphosphonates have been linked to increased rates of fractures of long bones such as the femur.25 Aside from increased risks of fractures, up to one-third of women given biphosphonates will have significant gastric distress, a small number (1 in 10,000) will get osteonecrosis (bone death) of the jaw or other bones, and an unknown number will develop generalized pain syndromes, or eye problems including blindness, or cardiac problems that may increase the risk of a stroke.26 The difficulty in knowing how many will suffer from these complications is that the company trial data is almost unbelievable—in the company trials that have been released the rates of gastric problems are the same on the active drug as for the placebo.
These risks might be worth running for those rare women who have a clinically established form of osteoporosis if there were good evidence that medication could help or help as much as getting fit. There is good evidence that factors such as levels of a person's physical fitness are better predictors of fractures than are bone densities, and good evidence that encouraging women to improve their fitness will reduce their fracture rates.27 But fitness clinics don't distribute articles with the results of controlled trials demonstrating such benefits of good exercise, whereas doctors are bombarded with articles in which an artful use of statistics appears to confirm a benefit to drug treatment, and women are subjected to ubiquitous advertisements on the benefits of the drug. And just as with the statins, for both doctor and patient prescribing a pill can seem easier than changing a lifestyle.
From the 1990s onward magazines for teenage girls began to feature discussions of osteoporosis. While none of these adolescents will have been the direct target of marketing by Merck and Procter and Gamble, still the change in culture these companies have created feeds through to all women, reinforcing the message in ads that are directly aimed at older women. Messages such as those in a Merck ad showing a very attractive woman apparently emerging from the bath, with a towel discreetly held to her upper chest but displaying enough of her body for the ad to be confident you will give a positive answer to its first question, “See how beautiful sixty can look?” This is followed by a further question, “See how invisible osteoporosis can be?” The sidebar then tells us that one in two women sixty and over have osteoporosis, and that while it may be invisible it leads on to broken bones and a dowager's hump. This Merck ad is not for Fosamax by name; instead it asks women to ask their doctor if a bone density test might be right for them.
The numbers that come from bone density scans and blood tests for cholesterol set up a new normality. If our numbers are abnormal, we almost immediately begin to feel at risk—dis-eased. And here we come to a key point. Whereas for centuries medicine had almost exclusively been about the treatment of diseases in the sense of biological disorders that posed an immediate threat to our lives, and still is in much of the world, Western medical practice now is increasingly about the management of risks—and this increasingly creates dis-ease.
When it came to treating diseases, epidemics aside, usually only a relatively small number of people had them. In the West much greater numbers are at risk as now defined. There are thousands of women with technical osteoporosis now for every one with clinical osteoporosis in the past, thousands with checklist depression now for every one with melancholia in the past, and entire populations with lipid abnormalities where hypercholesterolemia had been a rare disorder in the past.
In early twentieth century, we were rescued from the problems of a previous generation of celebrity drugs by the laboratory work of scientists and doctors from Robert Koch to Richard Cabot and others. Many physicians today remain confident that mapping of the human genome and other scientific developments will save us from the ravages of disease tomorrow. But there is a critical difference: the bacteriology of Koch alerted us to an external threat such as cholera, anthrax, or tuberculosis that we could mobilize against, whereas the mapping of the human genome promises to extend the riskiness of our universe. There are likely to be a much greater number of genetic markers for small degrees of risk than there will be genes for diseases. How do we mobilize against the uncertainties of the human body itself?
Furthermore, whereas the treatment of disease stops once the disease clears up, when it comes to managing risk, treatment potentially goes on forever. There is no natural endpoint except death itself. Where medicine once aimed at eliminating disease, now, for a pharmaceutical company, the trick is to persuade as many people as possible that they are at risk of a multitude of “diseases” and to convince doctors that these are all conditions that should be tested for and treated.
Insofar as being alive is risky, a health products market devoted to risk management has the potential to swallow up huge domains of our experience. Hitherto, sensibly or not so sensibly, we have often seen the risks in being alive in spiritual terms. But where before it was ascetics who attempted to control their bodies by fasting and punishment, now almost all ages and both sexes go to the same extremes in the name of health. Where once people in the West were said to have been born with an original stain, now we feel as though we have been born with a warranty—and if anything goes wrong we want to know who is to blame. Where once the focus was on living well in order to die well, now it is more likely to be on living well in order not to die, or at least to put off its eventuality as long as possible. There is a potential here not just for a health market, but for an ersatz religious universe. And if you're a drug company executive, there's opportunity here to sell not just pills but something closer to a sacrament.
In previous times we passed on a culture to our children embodied in fairy tales, folklore about health, national myths, and religious precepts, in which the life's risks were put in a larger context of meaning. Now an increasing part of what is transmitted centers on personal health for its own sake: figures for sugar and lipid levels, as increasing numbers of our children have diabetes or other dangerous metabolic states, or figures for peak respiratory flows as increasing numbers of young people
have asthma, or statistics on some chemical imbalance as increasing numbers are being treated for ADHD, depression, or anxiety. Not only is such a culture two-dimensional, it changes the very nature of human experience.
MEASURING DIS-EASE
The new focus on blood sugars, blood lipids, bone density, and the like has transformed the encounters between doctors and patients. But the change has not come simply from blood tests and other obvious measurements. It has also come from the use made of scales developed to measure aspects of behavior. These rating scales were needed in trials of antidepressants, tranquilizers, analgesics, hypnotics, drugs for sexual dysfunction, and other drugs used to modify behavior, for the same reasons as cholesterol levels are needed in statin trials and DXA scans in biphosphonate trials. In lieu of evidence that patients get up from their beds and walk, feel better again, and return to work, these rating scales produce numbers that go in the right direction on treatment and can be held up as evidence that the treatments are working.
Rating scales consist of a series of items that physicians enquire about and then score the responses. In the case of depression, these typically included sleep, appetite, energy and interest levels, suicidality, feelings of guilt, and agitation. One of the very first of these scales was the Hamilton Depression Rating Scale (HDRS), published in 1960.28 Although notionally credited to Max Hamilton, the scale was put together by Geigy to use in clinical trials of imipramine, their new antidepressant.
Many clinicians were initially skeptical of the merits of what appeared to be a basic checklist, not unlike the checklists in magazines, that the office receptionist could be easily trained to administer. Far from arguing against this clinical gripe, Hamilton apparently believed the main merit of the scale was that it facilitated the conduct of drug trials. Using the scale appeared to show imipramine and similar drugs, which increased appetite and sleep, worked compared to placebo. Clinicians could see there was a benefit with their own eyes, but rather than leave the matter to clinical judgment, the scale offered an apparently objective measure of effectiveness. But Hamilton was also aware that the wider use of such checklists might induce a substantial change in culture: “It may be that we are witnessing a change as revolutionary as was the introduction of standardization and mass production in manufacture. Both have their positive and negative sides.”29
Forty years later, there are few clinicians who can remember that even Hamilton thought that checklists such as these were an abstraction from the richness of clinical reality. The rating scales that were initially validated by clinical judgment are now being increasingly imported into clinical practice to invalidate clinical judgment, apparently in the belief that reducing variability in the clinical encounter will make that encounter more “scientific.” Whatever the merits of using a scale in a trial, it would not have made sense to Hamilton to approach every depressed patient the same way. Clinical encounters that use a rating scale will resemble each other much more than encounters with physicians where no rating scale is used. This might sound momentarily appealing—but clinical encounters based on a rating scale will also be difficult to distinguish from encounters with a receptionist trained to administer the scale, and this is not what we want from a doctor. We want the experience and discretion that goes with a clinical training rather than an encounter that ultimately could be delivered by computer.
Primary care physicians prescribe the greatest amount of psychotropic drugs these days and they are increasingly encouraged to administer depression or other behavioral rating scales when seeing patients. Many of the guidelines for prenatal care now advocate using anxiety and depression rating scales for all pregnant women. While this might pick up some women in need of treatment who would otherwise be missed, it will pick up a lot of women who aren't in fact depressed, but who because of temporary changes in sleep patterns or irritability may score in a zone of concern on rating scales one week but not the next. When the diagnosis is based on clinical judgment the estimate is that 3–4 percent of women may have a nervous disorder prenatally compared with estimates of 15–25 percent of women when the diagnosis is based simply on rating scale scores. These rating scale scores all too often translate into treatment, without further thought. As a result, the antidepressants have moved in less than ten years from being rarely used prenatally to being among the commonest drugs given in pregnancy—despite convincing evidence that they double the rate of birth defects and of miscarriages.30
Aware of this ambiguity in the value of rating scales—and the drug prescriptions that are likely to follow from their use—pharmaceutical companies now run symposia at major professional meetings aimed solely at introducing clinicians to rating scales. In the 1990s companies engaged in disease mongering by selling diseases such as social anxiety disorder or panic disorder in order to sell their drugs.31 In the decade 2000–2010 what we might call measurement mongering has succeeded disease mongering as the key promotional instrument. Thus Pfizer at the 2007 American Psychiatric Association meeting, for example, supported a symposium entitled “From Clinical Skills to Clinical Scales: Practical Tools in the Management of Patients with Schizophrenia.” The practical tools in question were rating scales whose items draw attention to ways in which the company's drug was superior to some others in the field. This is an ad in the form of a rating scale. It is just the same in its effects as makers of asthma drugs providing peak flow meters to doctors and Merck and Procter and Gamble providing DXA scanners, all the way back to the makers of antibiotics in the 1960s who provided thermometers to doctors.
Rating scales are not used just to sell diseases to doctors. Increasingly, companies disseminate rating scales directly to patient groups. Among the most strikingly successful maneuvers in this area has been to encourage patients with nervous problems to keep mood diaries. Getting patients to chart fluctuating emotional states has been a potent way to persuade both patients and their doctors that the patient has a bipolar disorder. In very short order, these methods have led to a boom in the diagnosis of bipolar disorder, with patients formerly seen as being depressed or anxious relabeled as bipolar and their treatment changed from one of the off-patent antidepressants to an on-patent mood stabilizer.
What has been lost since Hamilton's day is any sense that these rating scales are simply checklists. Far from being information rich, they are information poor. They may come with pompous, scientific-sounding names, but they have little more content than the checklists seen in periodicals like Vogue or Esquire that offer to map out aspects of our personal styles as lovers, socialites, or foodies. The main advantage likely to accrue from the use of these lists is to ensure that a number of questions that steer a doctor toward a particular prescription are checked off as asked. The allure to companies of these readymade questions in timelimited clinical exchanges is just this—these questions redefine clinical realities by pushing out what could be more probing and important questions that might lead away from drug treatments and toward efforts to modify lifestyles or change social situations. Increasingly, practice will be standardized—to the lowest common denominator.
When rating scales are imported into healthcare the clinical gaze risks being captured by those whose interests are served by the measurement technology. The rating scales add to company abilities to hypnotize clinicians and enfold them in a world defined by the marketing departments of pharmaceutical companies.
But it is not only the clinical gaze that is captured. The very nature of human experience can be redefined. Thus the creation of female sexual dysfunction (FSD) hinged on producing treatment changes on rating scales that include items such as clitoral numbness—because this is the kind of thing that can be tallied on a rating scale. Focusing on whether one's clitoris is numb or not, while making love, risks changing the entire experience of making love. It also creates a discontent, and perhaps focuses other discontents on this area, with a drug becoming the apparent answer to these discontents.32
The rating scale maps out the contours of
this new condition. The measurement and disease mongering of FSD has led in recent years to women in their twenties with three young children who have lost interest in sex being encouraged to think they should have their testosterone levels checked rather than considering that their problems may stem from the circumstances in which they find themselves.33 But aside from the women who may opt for low-dose testosterone as a result, the wider culture surrounding these issues has changed in ways that affect how all women perceive what is going on in them, in a manner analogous to the way the marketing of HRT up until recently persuaded a lot of women that being postmenopausal was to be diseased.
While rating scales, along with blood tests for cholesterol levels or bone scans for bone densities, do generate data, exclusive reliance on such data, an increasing temptation among harried doctors, leads to what I call informational reductionism. Critics have complained for decades about the biomedical reductionism that supposedly dehumanizes clinical exchanges. Reducing humans to their bodies—whether hormones or neurotransmitters or the mechanical actions of their heart— for these critics is not a fit way to treat people. But where an old-style medical encounter might lead to a focus on some important aspect of a patient's physical state, the upside of such reductionism has always been that the doctor might thereby pinpoint something that would help lift us out of some real disability or even save our life. In contrast, there is not a single benefit likely to accrue from this new informational reductionism whether embodied in lipid levels, bone density measurements, or rating scale scores—although the measurements are sold as empowering us.