From the "silent killer" to the "whispering disease": ovarian cancer and the uses of metaphor.

of the ravages of this insidious ... neoplasm ... than succumb to cervical and endometrial cancer combined”, and another reported in 1983 that “one woman dies from ovarian cancer every 50 minutes in this country”. 71 In the face of such discouraging facts, researchers sought new means of gaining con- trol over the disease. Important advances in surgery and chemotherapy were made, but their long-term impact was limited by the fact that a majority of cases were not diag- nosed until metastasis had occurred. 72 By the 1980s, more concerted attention was finally given to the long-neglected epidemiology of ovarian cancer, and some progress was made in identifying women at increased risk who might benefit from increased sur- veillance. 73 Much effort was devoted, as well, to development of mass screening techni- ques which might make earlier diagnosis possible for all patients, but results on that front were disappointing. 68 Way, op. cit., note 51 above, pp. 183–7. 69 Muriel L Newhouse, R M Pearson, J M Fullerton, E A Boesen and H S Shannon, ‘A case control study of carcinoma of the ovary,’ Br. J. Prev. Soc. Med., 1977, 31: 148–53, p. 148; M Steven Piver, Joseph J Barlow and Diane M Sawyer, ‘Familial ovarian cancer: increasing in frequency?’, Obstet. Gynecol., 1982, 60: 397–400, p. 399; Ralph C Benson, Handbook of obstetrics and gynecology , Los Altos, CA, Lange Medical Publications, 1983, p. 641; V Beral and M Booth, ‘Occurrence and etiology’, in Norman M Bleehen (ed.), Ovarian cancer, Berlin, NY, Springer Verlag, 1985, pp. 14–22, on p. 14. 70 Marchetti, op. cit., note 58 above, p. 370; M Steven Piver, ‘Epidemiology of ovarian cancer’, in M Steven Piver (ed.), Ovarian malignancies: diagnostic and therapeutic advances , Edinburgh, Churchill Livingstone, 1987, pp. 1–11, on p. 1; M Steven Piver with Gene Wilder, Gilda’s disease: sharing personal experiences and a medical perspective on ovarian cancer, Amherst, NY, Prometheus, 1996, p. 35. 71 J D Woodruff, ‘The pathogenesis of ovarian neoplasia’, Johns Hopkins Med. J. , 1979, 144: 117–20, p. 117; William M Rich, ‘Benign and malignant ovarian neoplasms’, in Ralph W Hale and John A Krieger (eds), A concise textbook of gynecology, Hyde Park, NY, Medical Examination Publishing, 1983, p. 297. 72 Rich, op. cit., note 71 above. 73 Margaret Booth and Valerie Beral, ‘The epidemiology of ovarian cancer’, in C N Hudson (ed.), Ovarian cancer, Oxford University Press, 1985, pp. 22–44, on p. 22. Patricia Jasen 500 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press The first risk factor to be systematically examined was low parity, although there was much disagreement over why it might predispose women to ovarian cancer. The debate centred on the question of whether “poor reproductive performance”, optional or other- wise, was in itself a source of risk, or whether low parity was to be interpreted as a sign of poor ovarian health which might lead to malignancy. 74 The New York ovarian cancer expert Hugh Barber declared in 1978 that: “Women at high risk usually have a long history of ovarian imbalance or dysfunction, including ... a tendency for sponta- neous abortion, infertility, and nulliparity, as well as an early menopause. The ovary is like a cam running off center.” 75 At the same time, other researchers emphasized that the protection afforded by pregnancy was probably the key factor, because unmarried and childless married women both had a higher incidence of the disease. 76 More studies followed and yet the question remained open, complicated by the fact that the known causes of infertility were so diverse. 77 Attention was also turned to another hypothesis, first put forward in 1971, that unin- terrupted or “incessant” ovulation might be an important factor in the aetiology of ovar- ian cancer. 78 This theory was compatible with studies suggesting that the contraceptive pill, which prevented ovulation, had a very significant protective effect. The U.S. Center for Disease Control, for example, estimated that “oral contraceptive use alone may have prevented up to 1700 cases of ovarian cancer” in 1982. 79 Some years later, the theory that frequent ovulation might predispose women to the disease provided an explanation for an emerging association between ovarian cancer and the use of fertility drugs. As women became more vocal about their right to know about potential risks associated with drugs and medical treatments, some cancer patients would charge that fertility clinics were negligent in not providing this information. 80 The question of a hereditary factor was also given serious consideration in 1980s. Instances of individual families with a high incidence of the disease had been mentioned in the late nineteenth century, and, from the 1930s onwards, studies of affected families grew slowly in number. One such study, published in 1950 by Amour Fiscus Liber, con- cerned ‘Ovarian cancer in mother and five daughters’. At the time, Liber was a lone voice calling for radical action based on his assumption of a genetic link. Although, in his words, he stopped short of advocating “eugenic limitation of breeding”, he recom- mended prophylactic oophorectomy for women in families thus affected, close monitor- ing from an early age, and the creation of agencies to keep records of all cases—even 74 J Donald Woodruff, ‘Premalignant and malignant disorders of the ovaries and oviducts’, in Martin L Pernoll (ed.), Current obstetric and gynecologic diagnosis and treatment, Norwalk, CT, Appleton and Lange, 1991, p. 974; Beral and Booth, op. cit., note 69 above, p. 17; Booth and Beral, op. cit., note 73 above, pp. 27–8. 75 Hugh R K Barber, Ovarian carcinoma: etiology, diagnosis, and treatment , New York, Masson, 1978, p. 98. 76 Newhouse, et al ., op. cit., note 69 above, pp. 152–3. 77 Susan Harlap, ‘The epidemiology of ovarian cancer’, in Maurie Markman and William J Hoskins (eds), Cancer of the ovary , New York, Raven Press, 1993, p. 83. 78 Booth and Beral, op. cit., note 73 above, p. 29. 79 Lloyd H Smith and Richard H Ol, ‘Detection of malignant ovarian neoplasms: a review of the literature. I. Detection of the patient at risk; clinical, radiological and cytological detection’, Obstet. Gynecol. Surv. , 1984, 39: 313–28, p. 314; Harlap, op. cit., note 77 above, p. 83. 80 Piver with Wilder, op. cit., note 70 above, pp. 42–4. For a patient’s account of the controversy over fertility drugs in the 1990s, see Liz Tilberis, No time to die, Boston, Little, Brown, 1998, pp. 40–5, 246–54. Ovarian Cancer and the Uses of Metaphor 501 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press suggesting that, if possible, autopsies be performed not only on patients but on all rela- tives, including males, in case “there is a recognizable phenotype of male carriers”. 81 During the 1950s and 1960s, the importance of a genetic link was widely dismissed, but by the 1970s the number of families on record was increasing rapidly, and there was speculation that the incidence of familial disease was actually growing. 82 In 1981, the first Familial Ovarian Cancer Registry in the US was established at Roswell Park Memorial Institute in Buffalo. Its goal was not merely to study the “mode of inheritance” but to examine relationships with breast and other cancers, as well as inter-connections with other suspected risk factors, including asbestos exposure. 83 That same year, a widely cited study reported that women with a mother or sister with the disease had an alarming eighteen-fold increase in risk, and during the 1980s there was much speculation upon the nature of the familial link, whether screening procedures should be introduced for high-risk women, and whether prophylactic oophorectomy should be recom- mended. 84 In the early 1990s, the BRCA1 and BRCA2 genes associated with both breast and ovarian cancer were localized, removing all doubt that women in certain families faced an extremely high level of hereditary risk. 85 Because the incidence of ovarian cancer was much higher in western industrialized countries than elsewhere, researchers also investigated cultural and environmental fac- tors. One of these was a possible association with asbestos exposure in the workplace and the use of talc (contaminated by asbestos until at least the mid-1970s) in personal hygiene. A 1960 study had suggested that women with asbestosis had a higher rate of ovarian cancer, and another published in 1971 directed attention towards the possible link with talc by reporting that, in the set of ovarian tumours which they examined, 75 per cent contained talc particles. 86 That study and others proposed that this factor might actually help to account for the rising incidence of ovarian cancer in the post- war decades. 87 Further investigations demonstrated that talc applied to the genital area could make its way very rapidly into the peritoneal cavity, and a case control study pub- lished in 1982 found that 42.8 per cent of the women with ovarian cancer had used talc regularly, compared with 28.4 per cent of the controls. 88 The association between 81 Amour Fiscus Liber, ‘Ovarian cancer in mother and five daugthers’, Arch. Pathol., 1950, 49: 280–90, pp. 289–90. Liber cites cases from 1877 onwards. 82 Romney, et al., op. cit., note 57 above, p. 1065; Piver, et al ., op. cit., note 69 above, p. 398. 83 Piver, et al ., op. cit., note 69 above, pp. 399–400; Piver with Wilder, op. cit., note 70 above, p. 46. 84 The study often referred to was G Hildreth, Jennifer L Kelsey, Virginia A Livolsi, Diana B Fischer, et al ., ‘An epidemiologic study of epithelial carcinoma of the ovary’, Am. J. Epidemiol. , 1981, 114: 398–405. And see Smith and Ol, op. cit., note 79 above, pp. 316–17; Booth and Beral, op. cit., note 73 above, pp. 30–1. 85 Barnaby D Rufford and Ian J Jacobs, ‘Identification and management of familial ovarian cancer’, in Henry C Kitchener, Jonathan A Ledermann and Andrew Miles (eds), Effective management of ovarian cancer , London, Aesculapius Medical Press, 2001, pp. 128–38. On the patient experience of hereditary risk and delays in its recognition, see Nina Hallowell, ‘Varieties of suffering: living with the risk of ovarian cancer’, Health, risk and society , 2006, 8: 9–26. 86 W J Henderson, C A F Joslin, K Griffiths and A C Turnbull, ‘Talc and carcinoma of the ovary and cervix’, J. Obstet. Gynaecol. Br. Commonw. , 1971, 78: 266–72. 87 Ibid., p. 271; Woodruff, op. cit., note 71 above, p. 120. 88 Howard C Jones III, Anne Colston Wentz and Lonnie S Burnett, Novak’s textbook of gynecology , 11th ed., Baltimore, Williams and Wilkins, 1988, p. 793; Daniel W Cramer, William R Welch, Robert E Scully and Carol A Wojciechowski, ‘Ovarian cancer and talc: a case-control study’, Cancer, 1982, 50: 372–6. Patricia Jasen 502 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press asbestos exposure, talc use and ovarian cancer remained controversial, however, because researchers continued to produce conflicting results. 89 In addition to the examination of potentia l risk factors, a second approach to the growing incidence of ovarian cancer was the search for a means of effectively screen- ing asymptomatic women—either entire po pulations, or only those deemed to be high risk. By the mid-1980s, three main screening t echniques were available; these were the manual pelvic examination, the ultrasoun d technology which had been developed dur- ing the previous decades, and the recently devised CA-125 blood test, which measured the level of a cancer antigen associated with ovarian cancer. 90 As already mentioned, many physicians viewed regular pelvic exa minations as an impor tant means of detect- ing potential cancers in asymptomatic wome n, but others increasingly opposed this strategy, arguing that pelvic examination was limited in sensitivity and did not yield “the desired percentage of early diagnose s”. Ovarian cancer was found in only about one in 10,000 such procedures, while, at the same time, about one examination in 125 revealed some sort of mass which, they pointed out, resulted in unnecessary anxi- ety and potentially hazardous surgery. 91 Ultrasound technology, meanwhile, held out the hope that masses not felt during examina tion could be detected through imaging, a n da l s ot h a tm a n u a l l yl o c a t i n gam a s sw o u ldb e“ s a f e r ”i fi tc o u l dt h e nb ea s s e s s e d without surgery. 92 Ultrasound offered a st ep forward but was found to be too inaccu- rate, and too costly, for general screening, while the CA-125 blood test was neither specific nor sensitive enough for regular use in asymptomatic women (though both came into use for monitoring women considered high risk). 93 Despite efforts to d e v e l o pa ne a r l yd e t e c t i o np r o g r a m m ew h i c hp a r a l l e l e dm a m m o g r a p h ya n dt h eP a p smear, no viable strategy emerged for ovarian cancer. 94 Meanwhile, during the second half of the t wentieth century, medical literature of various kinds re-confirmed the image of the “ silent killer”. “Ovar ian cancer, is unfor- tunately, very insidious and ‘silent’ in t erms of signs and symptoms”, reiterated Novak’s authoritative text in 1988, and s imilar language appeared in many other works.95 But, as the prospect of mass screening for ovarian cancer proved elusive, 89 Mark S Shahin and Joel I Sorosky, ‘Prevention and early diagnosis of ovarian cancer’, in Alberto Manetta (ed.), Cancer prevention and early diagnosis in women, Philadelphia, Mosby, 2004, pp. 249–66, on pp. 254–5. 90 On the development of ultrasound, see S Levi, ‘The history of ultrasound in gynecology 1950–1980’, Ultrasound Med. Biol. , 1997, 23: 481–552. 91 Walter J Burdette, Cancer: etiology, diagnosis, treatment, New York, McGraw-Hill, 1998, p. 166; Rich, op. cit., note 71 above, p. 297. 92 Smith and Ol, op. cit., note 79 above, p. 322. 93 Burdette, op. cit., note 91 above, p. 166; Marilyn F Vine, Roberta B Ness, Brian Calingaert, Joellen M Schildkraut and Andrew Berchuck, ‘Types and duration of symptoms prior to diagnosis of invasive or borderline ovarian tumor’, Gynecol. Oncol., 2001, 83: 466–71, p. 466; Clare Bankhead and Joan Austoker, ‘Women’s cancer screening: cervical, breast, and ovarian screening’, in Deborah Waller and Ann McPherson (eds), Women’s health , Oxford University Press, 2003, p. 484; Barbara A Goff, Lynn S Mandel, Cindy H Melancon and Howard G Muntz, ‘Frequency of symptoms of ovarian cancer in women presenting to primary care clinics’, JAMA, 2004, 291: 2705–12, p. 2710. On a patient’s experience with the unreliability of ultrasound, see Barbara R Van Billiard, A feather in my wig: ovarian cancer cured , Portsmouth, NH, Peter E Randall, 1998, pp. 1–4. 94 Goff, et al ., op., cit., note 93 above, p. 2710; Shahin and Sorosky, op. cit., note 89 above, p. 259. 95 Jones III, et al ., op. cit., note 88 above, p. 793. For other examples during the post-war decades, see Robert James Crossen and Ann Jones Campbell, Gynecologic nursing , St Louis, C V Mosby, 1956, pp. 136–7; Elizabeth Parker, The seven ages of woman, Baltimore, Johns Hopkins Press, 1960, Ovarian Cancer and the Uses of Metaphor 503 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press some voices were raised in favour of giving m ore attention to the n on-specific symp- toms which women could detect themselves, and of questioning the very use of the term “silent killer”. “Evidence that curren t screening techniques reduce mortality is lacking”, wrote one team. “Therefore, symp tom recognition is important in the detec- tion of ovarian cancer.” 96 Due to “problems with sensitivity and specificity”, decided another, “we have to rely on the woman and her initiative in order to achieve an early diagnosis.” 97 From the “Silent Killer” to the “Whispering Disease” Increasing attention to ovarian cancer symptoms coincided with a much wider women’s health movement in the 1980s and 1990s. Its objectives were many, but a cen- tral goal was to transform the clinical encounter through the legitimization of patients’ own experiences of health and illness. For the nascent ovarian cancer movement, much of the focus was on reducing the risk of wrong or delayed diagnosis through edu- cating both physicians and the public about the common symptoms of “the disease that whispers”. This campaign would eventually lead to collaboration between patient acti- vists and members of the research community and, by the early twenty-first century, to official recognition of a pattern of what would now be termed “early” symptoms, accom- panied by a widening critique of the “silent killer” metaphor. A start was made in 1978 when Hugh Barber revived Stanley Way’s plea of nearly three decades before, declaring it was time “to change the generally accepted notion that there are no early symptoms” of ovar ian cancer. Like Way, Barber observed that women’s complaints of increased girt h and gastro-intesti nal problems were rou- tinely dismissed or wrongly diagnosed. “Al lt o oo f t e n ” ,h ew r o t e ,“ t h ep a t i e n ti sc o n - sidered a middle-aged crock who goes to too many cocktail parties and eats too many hors d’oeuvres.” 98 A year later, the South Dakota physician Brooks Ranney used patients’ records to tabulate the symptom s they had experienced and confirmed that most patients, including women with stage 1 and 2 disease, who therefore had a much better prognosis, had noticed symptoms over a period of two weeks to ten years. 99 In 1985, an Iowa team collected information from patients rather than from clinical records, and found that “[i]n s harp contrast to clinical perceptions and p. 554; Marchetti, op. cit., note 58 above, p. 369; Beacham and Beacham, op. cit., note 57 above, p. 291; Romney, et al ., op. cit., note 57 above, p. 1067; Woodruff, op. cit., note 71 above, p. 117; Sandra L Tyler and Gail M Woodall, Female health and gynecology across the lifespan , Gowie, MD, Robert J Brady, 1982, p. 226; Benson, op. cit., note 69 above, pp. 640–1; Mary Daly and G Iris Obrams, ‘Epidemiology and risk assessment for ovarian cancer’, Semin. oncol. , 1998, 25: 255; Jo Ann Rosenfeld, ‘Ovarian cancer and ovarian masses’, in Jo Ann Rosenfeld (ed.), Handbook of women’s health: an evidence-based approach , Cambridge University Press, 2001, pp. 333–48, on p. 334. 96 Vine, et al ., op. cit., note 93 above. 97 C Wikborn, F Pettersson and P J Moberg, ‘Delay in diagnosis of epithelial ovarian cancer’, Int. J. Gynecol. Obstet. , 1996, 52: 263–7, p. 266. 98 Barber, op. cit., note 75 above, p. 97. Barber may well have inspired other researchers to turn their attention to this issue for, according to his obituary, he was “internationally renowned for his seminal work in ovarian cancer”. The New York Times , 29 Dec. 2006, http://query.nytimes.com/gst/fullpage. html (accessed 2/6/2009). 99 Brooks Ranney and M I Ahmad, ‘Early identification, differentiation, and treatment of ovarian neoplasia’, Int. J. Obstet. Gynecol. , 1979, 17: 209–19. Patricia Jasen 504 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press previous research ... three fourths of women with locally staged tumors of the ovary” had symptoms ranging from swelling to fatig ue, pain, problems wi th urination, indi- gestion, irregular bleeding, shortness of b reath, and bowel changes. Nearly half of the women had dismissed these discomforts as “not serious”, especially those under forty and over fifty. They concluded that l ocalized cancer did have symptoms, and that helping women to recognize them was a n important public health priority. 100 Phy- sicians, as well, needed to know what they we re, and a Swedish team pointed out that, “[h]owever vague the symptoms, there are some that should alert the clinician to the possibility of ovarian cancer”. 101 That call was repeated at intervals during the 1990s. Researchers continued to observe the long delays that separated the onset of symptoms from correct diagnosis and the chal- lenge of finding a doctor “who is familiar with the symptoms of ovarian cancer”. 102 Their efforts attracted less attention than they might have hoped, and physicians’ uneven knowledge of the disease was the subject of a pair of studies conducted in 1999. The first surveyed primary care physicians and achieved a response rate of just over one half. Within this group, the researchers reported a good deal of variation in respondents’ abil- ity to identify risk factors correctly, while the accurate identification of symptoms ranged from 60 per cent for weight gain to 94 per cent for ascites (abdominal fluid). 103 An odd feature of this first study was that the authors’ list of six symptoms omitted any reference to the common signs of indigestion, bloating, and changes in bowel habits. The second study, however, which measured gynaecologists’ perceptions, did include abdominal bloating and altered bowel function (but excluded indigestion), and a strong majority of respondents recognized these symptoms. Only 62 per cent identified the use of fertility drugs as a potential risk factor, however, and 71 per cent identified a previous history of breast cancer. 104 The contrast between physicians’ sometimes limited knowledge and their seemingly unlimited authority was of major importance to women’s health activists in the 1980s and 1990s. This growing movement questioned the assumption that lay knowledge was necessarily inferior to biomedical knowledge and protested the dismissal of women’s intimate experience of their bodies. Activists sought to recover “the voice of the subject” and to recognize the role of intuition in relation to health and illness, despite the difficul- ties inherent in defining that concept. 105 These concerns would have particular resonance for the ovarian cancer movement, which—while hampered by the fact that many potential 100 Elaine M Smith and Barrie Anderson, ‘The effects of symptoms and delay in seeking diagnosis on stage of disease at diagnosis among women with cancer of the ovary’, Cancer, 1985, 56: 2727–32. 101 Folke Flam, Nina Einhorn and Kerstin Syovall, ‘Symptomatology of ovarian cancer’, Eur. J. Obstet. Gynecol. Reprod. Biol. , 1988, 27: 53–7, p. 53. 102 Gamal H Eltabbakh, Pramila R Yadev and Ann Morgan, ‘Clinical picture of women with early stage ovarian cancer’, Gynecol Onccol. , 1999, 75: 476–9, p. 479; Wikborn, et al ., op. cit., note 97 above, p. 266. 103 Ross E Gray, P Chart, J C Carroll, M I Fitch and D Cloutier-Fisher, ‘Family physicians’ perspectives on ovarian cancer’, Cancer Prev. Control, 1999, 3: 61–7, pp. 62, 64. 104 Margaret I Fitch, R E Gray, A Covens, Thomas G Franssen, et al ., ‘Gynecologists’ perspectives regarding ovarian cancer’, Cancer Prev. Control, 1999, 3: 68–76, pp. 71–2. 105 Laura K Potts, ‘Introduction: ‘Why ideologies of breast cancer? Why feminist perspectives?’ in Laura K Potts (ed.), Ideologies of breast cancer: feminist perspectives , London, Macmillan, 2000, pp. 1–11, on pp. 2–3; Jennifer Fosket, ‘Problematizing biomedicine: women’s contruction of breast cancer knowledge’, in ibid., pp. 15–36; Sara M Morris, ‘Lumps in the breast: negotiating risks Ovarian Cancer and the Uses of Metaphor 505 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press activists did not survive many years after treatment began—was intensely motivated by the need to improve the possibility of earlier diagnosis. Like many other aspects of the women’s health movement, ovarian cancer activism was strongly influenced by what Laura Potts described in her study of breast cancer nar- ratives as “a dominant culture of revelation, disclosure, and the making of testimony”. 106 The disease story is an old genre, but during the late twentieth century it became a powerful means of giving voice to the personal experience of cancer and of furthering the goals of the women’s health movement. 107 This sense of purpose was shared by writers of ovarian cancer narratives and, as was the case with breast cancer, the process began with the publication of testimonies written by well-known women such as the NBC correspondent Betty Rollin (who wrote about her mother’s illness and death from ovarian cancer), the comedian and actress Gilda Radner, Cosmopolitan editor Barbara Creaturo, and Liz Tilberis, editor-in-chief of Harper’s Bazaar . 108 Radner revealed that she unknowingly had most of the risk factors for ovarian cancer, including a strong family history and the use of fertility drugs. She did not recognize her own symptoms (“my stomach felt bloated and hard”), was diagnosed with stomach pro- blems by her gynaecologist and gastroenterologist, and spent months seeking an explana- tion for her increasing painful condition before finally being told she had ovarian cancer, stage 4. Her physician and family friend Steven Piver recalled that, at the time of her death in the spring of 1989, he assumed that the publicity surrounding her ordeal would mean that “the days of no newspaper or magazine articles or television specials on ovar- ian cancer were over”. 109 Radner’s death was certainly a turning point, with both Piver and her husband, Gene Wilder, taking up the cause of early detection, but the struggle for public awareness would continue. Many women who were not in the public eye were also motivated to publicize their experiences, whether in books, newspaper articles, or online. By the turn of the century, ovarian cancer patients by the hundreds were telling their stories on the internet where, as for many diseases, websites were being created which provided ordinary people with a forum for sharing their perspectives on the experience of cancer. On these sites, entries could be very brief and still serve some of the same purposes as other cancer nar- ratives. As Shani Orgad writes in Storytelling online , the process of writing “allows the author to make sense of her experience; to organize people, events, and information that she encountered, into a coherent framework of meaning”. 110 Using such narratives as historical sources obviously has limitations; as Potts says, disease stories tend to have “an aura of authenticity” when in reality they are one person’s version of events, possibly after a cancer diagnosis’, Health, Risk and Society , 1999, 1: 179–94, pp. 184–5. 106 Laura K Potts, ‘Publishing the personal: autobiographical narratives of breast cancer and the self’, in Potts (ed.), op. cit., note 105 above, pp. 98–127, on p. 98. 107 Ibid., pp. 99–103. 108 Betty Rollin, Last wish , New York, Linden Press, 1985; Gilda Radner, It’s always something , New York, Simon and Schuster, 1989; Barbara Creaturo, Courage: the testimony of a cancer patient , New York, Pantheon, 1991; Tilberis, op. cit., note 80 above. 109 Radner, op. cit., note 108 above, pp. 53, 58, 72; Piver with Wilder, op. cit., note 70 above, pp. 27–9, 42, 67–9, 20. 110 Shani Orgad, Storytelling online: talking breast cancer on the internet , New York, Peter Lang, 2005, pp. 36–7. Patricia Jasen 506 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press related in a time of crisis. 111 As a means of entry into that experience of crisis, however, they are a valuable resource. Ovarian cancer narratives are both similar to and different from most other types of cancer stories. A shared feature is the motif of “the enterprising self”, as Orgad calls it, through which the patient is cast as the protagonist embarking on a challenging jour- ney.112 In ovarian cancer narratives, however, the theme of “the enterprising self” often emerges most powerfully during the quest for a diagnosis, when the obstacle to be over- come, through courage and persistence, is not yet the disease itself but medical profes- sionals who do not recognize the symptoms of ovarian cancer or seem unwilling to listen to patients’ complaints. All such accounts are necessarily retrospective; they look back, after a diagnosis has been made, to the symptoms which preceded it and the false starts which were made in identifying the disease. In writing such narratives, some patients re called that they were n ot aware that ovar- ian cancer was called the “silent killer” un til after their disease was recognized, but they later identified the term as a factor in their delayed or incorrect diagnoses. One such story was that of Ayala Miron, the editor of a book titled Ovarian cancer jour- neys. She was diagnosed in 2000 after several years of reporting symptoms to her doc- tors and two trips to the emergency ward, wh ere she was diagnosed with a gallbladder attack on the first visit and diverticulitis on the second. 113 “As it turned out”, wrote Miron, my health care providers had completely misdiagnosed my symptoms. They didn’t know enough about ovarian cancer and did not suspect that my complaints were serious. After my ovarian cancer diagnosis, I realized that this disease caused the symptoms I felt. I also learned that many health care providers mistakenly consider ovarian cancer “a silent disease”. My symptoms, over a number of years, taught me differently. 114 Miron attributed health care professionals’ apparent lack of awareness to the influence of the “silent disease” image but also suggested that she, herself, failed to identify the symptoms she was experiencing because she was uninformed. The Johns Hopkins Pathology Ovarian Can cer Website provided a forum for women to write about their personal experiences with ovarian cancer in the early twenty-first century, and it became a particularly rich s ource of patient perceptions regarding the process of diagnosis. Stories of long del ay were legion, and a repeated theme was that both physicians and patients must be made aware of the symptoms which may sig- nal ovarian cancer. For example, Amy Chaik lin described “the laparoscopic discovery of ovarian cancer after 13 years of suffering”, while Becky Bennett recalled experienc- ing swelling and discomfort in the early 1990s , noticed a “lop-sided” abdominal enlar- gement by 1994, began to have problems w ith urination in 1999, and was diagnosed with a ten-pound tumour in 2001. 115 A woman signing herself as “Barbara”, in a tes- timony reminiscent of Stanley Way’s warnin g fifty years earlier, wrote of “popping 111 Potts, op. cit., note 106 above, p. 99. 112 Orgad, op. cit., note 110 above, p. 62. 113 Ayala Miron, ‘Symptoms are opportunities’, in Ayala Miron (ed.), Ovarian cancer journeys , Lincoln, NE, iUniverse, 2004, pp. 11–17, on p. 14. 114 Miron, ‘Foreword’, in ibid., pp. xiii–xiv, on p. xiii. 115 Johns Hopkins Pathology, Ovarian Cancer, Community, Personal Stories, http://ovariancancer. jhmi.edu/menu community.cfm, Amy Chaiklin, Becky Bennett (all stories accessed 2 June 2009). Ovarian Cancer and the Uses of Metaphor 507 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press Tum’s [ sic] and over-the-counter gas medicati ons like crazy” and being diagnosed with acid reflux, while “Chris Y.” wrote o f being told for three years that she had endometriosis and was then treated for a kidney infection before her diagnosis. 116 Donna McNulty related that h er swollen stomach was attributed to peri-menopausal weight gain; her bloating and nausea were treated as acid reflux; and her constant need to urinate was diagnosed as a bladde r infection—all common occurrences in women, but symptoms which, seen together, c omprise a pattern typical of ovarian can- cer.117 Judy Lidgate experienced the usual symptoms and was told she was suffering from depression, while Karen Leonard’s in tense pain was attributed to gallstones or a parasite. 118 An especially common misdiagnosis was irritable bowel syndrome, or IBS, and women sometimes blamed themselv es for not questioning their diagnosis. Augusta Gluck admitted, “I regret that I acce pted the diagnosis of irritable bowel syn- drome and did not investigate further ... But I had seen my internis t, my gynecologist, and a gastroenterologist. No one even me ntioned that my symptoms might have been ovarian cancer and my life threatened by it.” 119 Often employing the term “whispering disease” brought into use by the ovarian cancer movement, such testimonies addressed the relationship between the two principal meta- phors considered in this paper. They argued that ovarian cancer is not “silent” but that women must “listen” to their bodies in order to recognize the signs of “the disease that whispers”. They urged others to put aside their fears of disapproval, trust their own per- ceptions and intuitions, and—shifting their use of the word “listen” to its more literal sense—demand their doctors’ attention: “Don’t just listen to your body (we all know or have heard that ovarian cancer whispers), make sure the doctor is listening, too.” 120 The notion of “the disease that whispers” itself came under criticism, however, for it lacked resonance for women whose symptoms had become intense well before they were diagnosed. As one woman declared in a newspaper interview, “It’s such a whimpy, cop-out term. I hate it. It’s not a whispering disease. It’s a shouting disease.” 121 Another woman, a long-term survivor of stage 4 cancer, told of how by the time a diagnosis was reached “the ‘disease that whispers’ was screaming out to her”. 122 The majority, none the less, focused on the need to heed the earlier, subtler symptoms of ovarian cancer, and various permutations of “the whispering disease” metaphor gained wide currency in the ovarian cancer movement. 123 In spite of the efforts of the ovarian cancer movement, many women’s health manuals and menopause guidebooks, written by women’s health advocates and physicians, were slow to reflect need for more information about symptoms of the disease. In part, this reticence may have reflected other priorities of the women’s health movement. The 116 Ibid., Barbara, Chris Y. 117 Ibid., Donna McNulty. 118 Ibid., Judy Lidgate, Karen Leonard. 119 Ibid., Augusta Gluck. And see Amy Chaiklin, Vanessa Marshall, Jan Witsoe, Kate “jemakri” beckman ( sic). 120 Ibid., Beverley; and see Diane Paul, Diane McNulty. 121 Freda Ariella Muscovitch, quoted in Gabor Mate, ‘Ending the deadly silence’, Globe and Mail , 2 Oct. 2001, p. R5. 122 Marlene Eisner, ‘The disease that whispers’, http://thesuburban.com (accessed 1/6/2009). 123 Jane Brody, ‘The deadly whispers of ovarian cancer’, New York Times , 2 Oct. 2001, online. http:// www.nytimes.com/2001/10/02/health/personal- health-the-deadly-whisper-of-ovarian-cancer.html (accessed 1/6/09). Patricia Jasen 508 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press prophylactic removal of healthy ovaries when hysterectomies were performed was still common, and one way to counter that practice was to downplay the danger of ovarian cancer. Lonnie Barbach’s The pause , for example, assured readers that ovarian cancer is a sufficiently rare disease that the ovaries “should not be removed preventatively”. 124 Even in the early twenty-first century, most health manuals aimed at middle-aged women said little or nothing about ovarian cancer symptoms. In her Menopause and hormone book, Dr Susan Love passed lightly over the complexities of ovarian cancer diagnosis with the simple comment that there is “no good way to detect it early”, while Ivan K Strausz’s You don’t need a hysterectomy described the early symptoms of the disease as “entirely inconspicuous” and remained highly pessimistic about the prospect of timely diagnosis. 125 And yet, this neglect was far from uniform, for the Boston Women’s Health Book Collective had already begun, as early as the 1984 edition of Our bodies ourselves , to detail the “warning signs” of ovarian cancer which, they cautioned, “are frequently dismissed merely as ‘stress’ or nerves’”. These symptoms included “indigestion, gas, constipation or diarrhea, loss of appetite or weight, a feeling of fullness, lower abdominal discomfort or pain, frequent urination, fatigue, backache, nausea, vomiting, nonmenstr- ual vaginal bleeding, enlargement or bloating of the abdomen or an unusual growth or lump”. Persistence of such symptoms, they advised, called for “a thorough physical eva- luation for ovarian cancer” by means of a symptom review, family history, pelvic and rectal examination, CA-125 blood test, and ultrasound. 126 This disparity within the pop- ular women’s health literature shows that information about symptoms and diagnostic procedures was available to the public, even if many authors chose not to include it in their works. “Official” Recognition of Ovarian Cancer Symptoms By the late 1990s, a growing body of rese arch was providing more and more evi- dence that there was, indeed, a pattern of early warning signs of ovarian cancer which should be more widely publicized and acte d upon. The growing ovarian cancer move- ment, meanwhile, lobbied for further resea rch initiatives, and, on occasion, worked with members of the research community to document ovarian cancer symptom pat- terns. One leading activist was Cindy Melan con, a registered nurse living in Amarillo, T e x a s ,w h ow a sd i a g n o s e di n1 9 9 2a n ds t a r t e dan e w s l e t t e r ,Conversations ,a sa forum for women with ovarian cancer and a s a vehicle for disseminating informa- tion. 127 A seminal moment in the relationship between lay activists and professional researchers came in 1998, when Barbara Goff, a gynaecologic oncol ogist from Seattle, 124 Lonnie Barbach, The pause: positive approaches to menopause , New York, Penguin, 1995, p. 99. 125 Dr. Susan Love’s menopause and hormone book, New York, Three Rivers Press, 2003, p. 164; Strausz, op. cit., note 3 above, p. 303. 126 Boston Women’s Health Book Collective, Our bodies, ourselves , New York, Touchstone, 1984, p. 627. 127 Johns Hopkins Pathology website, op. cit., note 115 above, Cindy Melancon. Ovarian Cancer and the Uses of Metaphor 509 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press Washington, met Melancon at a conference for the lay public. As Goff described the meeting: Cindy and other survivors challenged the notion that ovarian cancer is a silent disease. After listen- ing to these survivors’ exceptional stories, I decided to team up with Cindy so that we could eval- uate symptoms and early diagnosis in a scientific manner. During medical school, residency, and even gynecologic oncology fellowship I had been taught that ovarian cancer was a silent disease and so initially I was not optimistic that our studies would yield new information. 128 Although a number of researchers were already challenging the language long used to describe ovarian cancer, the findings of Goff and her colleagues would eventually prove to be a turning point. Their first study was published in the American Cancer Society’s journal Cancer in 2000. It was a retrospective study based on information collected from ovarian cancer patients through a survey sent to subscribers of Melancon’s newsletter, and the authors concluded that the great majority of women diagnosed at an early stage experienced symptoms. They analysed the reasons for delays in detecting the disease, ranging from women’s own ignorance of symptoms to dismissive attitudes on the part of some physicians and the failure to perform pelvic examinations or order tests. While acknowledging the possibility of both selection bias (the women were Melancon’s sub- scribers) and recall bias (they were recalling symptoms after receiving their diagnoses), Goff’s team concluded that “women with ovarian carcinoma do have symptoms in con- trast to what is stated in most textbooks and taught in most medical schools”. 129 A sec- ond study responded to the problem of bias and also addressed a concern voiced by primary care physicians, which was that the non-specific symptoms of ovarian cancer are experienced by a great many women and can hardly be considered a guide to early diagnosis. By tabulating symptoms experienced during the past year by women about to undergo surgery for a pelvic mass compared with two control groups attending primary care clinics, the team found that, while women commonly reported at least one of the symptoms associated with ovarian cancer, those eventually diagnosed with the disease had much more severe, frequent and varied symptoms, even compared with women suffering from IBS. They concluded that their study “adds further evidence that ovarian cancer is not a silent disease”. 130 Other researchers pursued the same question, and their combined work eventually led to a public statement released in June 2007 by the American Cancer Society, the Gynecologic Cancer Foundation, and the Society of Gynecologic Oncologists announcing that, for the first time, a “national consensus” had been reached in the US regarding early signs of ovar- ian cancer. 131 The signs were listed as “bloating, pelvic or abdominal pain, trouble eating or feeling full quickly” and “urinary symptoms, such as urgent or frequent feelings of needing 128 Barbara Goff, ‘Introduction’, in Miron (ed.), op. cit., note 113 above, pp. xv–xvii, on p. xv. 129 Barbara A Goff, L Mandel, H G Muntz and C H Melancon, ‘Ovarian carcinoma diagnosis: results of a national survey’, Cancer, 2000, 89: 2068–75. 130 Goff, et al ., op. cit., note 93 above, pp. 2705–12. 131 ‘Ovarian cancer has early symptoms’, op. cit., note 26 above. Other studies included S H Olson, L Mignone, C Nakraseive, T A Caputo, et al ., ‘Symptoms of ovarian cancer’, Obstet. Gynecol. , 2001, 98: 212–17; Barbara P Yawn, Brigitte A Barrette and Peter C Wollan, ‘Ovarian cancer: the neglected diagnosis’, Mayo Clin. Proc. , 2004, 79: 1277–82; Lloyd H Smith, C R Morris, S Yasmeen, A Parikh-Patel, et al., ‘Ovarian cancer: can we make the clinical diagnosis earlier?’, Cancer, 2005, 104: 1398–1407. Patricia Jasen 510 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press to go”.132 The news media quickly spread the word, and Goff was one of the main spokes- people called upon to interpret these findings to the public. “There’s been this myth about ovarian cancer being silent and people saying there’s nothing you can do about it”, Fox News reported her as saying, “well, that’s simply not true anymore.” 133 In an interview with Jim Lehrer on PBS, the American public television network, Goff addressed the old problem of the non-specificity of symptoms by explaining that “it’s not simply just having the symptom that is important”, but that the symptom is new, persistent, and increases in severity. 134 “The majority of time”, she advised, such symptoms will not signify ovarian cancer, “but it’s just something that should be considered.”135 She recommended that persis- tent symptoms, when reported by patients, should lead to a pelvic examination, including “a recto-vaginal exam so that the ovaries can be appropriately felt”, followed by ultrasound and a CA-125 blood test, if warranted. If the ultrasound detected a mass and the blood test revealed an elevation of a tumour marker that is common with ovarian cancer, the patient should be referred to a specialist in gynaecologic cancers. 136 Debbie Saslow, director of breast and gynaecologic cancer at the American Cancer Society, was also widely quoted, and spoke to the still relevant concern over causing more women to undergo unnecessary and hazardous surgery. She revealed that the ACS still had reservations about the recom- mendations, due to the fact that “[w]e don’t have any consensus about what doctors should do once the women come to them”. 137 None the less, leading members of the cancer research community stood firmly behind the public statement. Columbia University’s direc- tor of gynaecologic oncology, Thomas J Herzog, commented, that “By no means do we want this to result in unnecessary surgery. But I would not expect that to occur in the vast majority of cases.” He also saw a need for physicians to reverse their old approach and discuss the existence of early symptoms with their patients so that women might become “more pro- active” in recognizing them. The gynaecologic oncologist Carol Brown, at Manhattan’s Memorial Sloan Kettering Cancer Center, agreed that “[t]his is something that women themselves can do”. 138 While the June 2007 announcement did not suddenly produce a consensus within the medical community regarding the significance of early symptoms, researchers would continue to investigate the role they might play in improving ovarian cancer diagnosis.139

This article has examined ovarian cancer’s long association with the metaphor of the “the silent killer” and has traced the process by which that metaphor, and the understand- ing of ovarian cancer symptoms which it signified, were eventually challenged in the 132 ‘Ovarian cancer has early symptoms’, op. cit., note 26 above. 133 Associated Press, ‘Doctors identify early symptoms of ovarian cancer’, FoxNews.com, 14 June 2007, http://www.foxnews.com/story/ 0,2933, 281653,00.html (accessed 1/6/2009). 134 ‘Doctors identify possible early signs of ovarian cancer’, 13 June 2007, http://www.pbs. org/newshour/bb/health/jan-june07/cancer 06–13. html (accessed 2/6/2009). 135 Denise Grady, ‘Symptoms found for early check on ovarian cancer’, New York Times , 13 June 2007, http://nytimes.com/2007/06/13cancer.html (accessed 2/6/2009). 136 Ibid. 137 Ibid. 138 Ibid. 139 B A Goff, L S Mandel, C W Drescher, N Urban, et al ., ‘Development of an ovarian cancer symptom index’, Cancer, 2007, 109: 221–7. Ovarian Cancer and the Uses of Metaphor 511 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press pages of medical journals and through the voices of ovarian cancer activists. It has attempted to show that, while the use of metaphorical language in medical discourse is inevitable, metaphors arise within particular contexts and may outlive their utility in expressing medical realities. In the early twentieth century, within the context of the “war” on cancer, the adoption of the “silent killer” metaphor in medical texts reflected the fact that the disease almost always progressed to an advanced stage before it was diagnosed and was almost always fatal. These factors also supported its exclusion from public information campaigns and popular medical literature aimed at women read- ers. From mid-century onwards, a small number of physicians openly questioned the accuracy of this metaphor and the medical assumptions which it signified, but only dur- ing recent years were the existence of early symptoms and their possible role in diagnosis widely and publicly addressed in both the lay and medical press. The association of the “silent killer” metaphor with ovarian cancer was exceptionally tenacious, and it arguably played a role in diverting attention away from systematic attention to symptoms which were later deemed relevant by a growing number of researchers. This history provides support for the conclusion that medical metaphors do have a role in defining “notions of reality” and therefore deserve close scrutiny. 140 140 Lupton, op. cit., note 17 above, p. 59. Patricia Jasen 512 https://doi.org/10.1017/S0025727300000521 Published online by Cambridge University Press