In February 1956, Pincus approached Edris Rice-Wray, medical doctor of the Associacion pro Bienestar de los Familias in San Juan (the Puerto Rican Family Planning Association) about the possibility of undertaking progestin trials in their birth control clinics. Rice-Wray, who held an appointment at the University of Puerto Rico and was medical director of the associacion as well as director of the US Public Health Field Training Center in Rio Piedras, seemed the ideal person to assist Pincus. She was experienced in the organization of family planning programs and had all the contacts necessary to gain access to possible test locations. Rice-Wray was willing to cooperate and agreed to supervise a series of prolonged field trials with Pincus’s compounds that would be carried out under the aegis of the Family Planning Association. Rice-Wray was to be in charge of the fieldwork which included

recruiting cases, distributing the contraceptive, monitoring the subject’s reaction, and collecting the required data. Pincus and Rock retained control over the research design and provided facilities for laboratory and other analyses that could not be done in Puerto Rico.

(Ramirez de Arellano and Seipp 1983:113)

Pincus thus found everything he needed at the Family Planning Association.

The first field trial was organized in the new public housing development project in Rio Piedras. The superintendent of this housing project, who was sympathetic to birth control and a person familiar to the tenants, agreed to cooperate in the organization of the trial and provided Rice-Wray and her team with records of the families. She subsequently selected the first set of 130 subjects, all women who had previously been given conventional barrier contraceptives at the clinic of the associacion (McLaughlin 1982:129; Maisel 1965:130-131). The screening of women for the trial consisted of a complete physical examination, carried out by Garcia, to ensure that the women were in good health (Maisel 1965: 131). Garcia also set up the meticulous data- keeping system (Garcia in Anonymous 1978:47).

This test location seemed to be the right one. The women at Rio Piedras were eager to participate in the project, even to such an extent that Rice – Wray had to create a waiting list (McLaughlin 1982:129). Many of these women had many children despite some use of other available contraceptives (Vaughan 1972:49). The crucial question now was how these women could be motivated to participate in the trial for a longer period. All previous tests had been of short duration: no woman had taken the progestins for longer than a few months. The field trials had to answer questions about the effects of long-term use of the new contraceptive (Vaughan 1972:41). The project would be successful only if women would not drop out too early. How could these women be turned into reliable producers of data?

The way in which the Rio Piedras trial was organized shows the strategies that Pincus and Rock used to discipline women to meet the requirements of their tests. The first requirement was that the women had to learn to follow the relatively complicated instructions of “one tablet a day, beginning on day 5 of the menstrual cycle until one vial of 20 tablets was consumed, i. e., through day 24 of the cycle” (Pincus 1958: 1, 335). Rice-Wray, in consultation with Pincus and Rock, assigned the responsibility to ensure a strict adherence to this regime of medication to a team of trained social workers. In the report of this trial, published in the American Journal of Obstetrics and Gynecology in 1958, Pincus described this regime:

A schedule of visits by a trained social worker was arranged so that in every medication cycle each subject was seen shortly after she should have taken the last tablet. Initially only one vial was distributed to each woman. This was replaced on the social worker’s visit. Since, in a number of instances, the housewives were not at home when she called, it was decided, after a few months, to leave two vials with each subject so that the continuity of the regime of medication (day 5 through day 24) might not be broken.

(Pincus 1958:1,335)

During the visits of the social worker, the women were asked to cooperate in interviews as well.

At each consultation, information was elicited concerning the length of the menstrual cycle, the occurrence of side effects, the frequency of coitus, and the number of missed tablets. A rough check, in some

instances, on the number of tablets omitted was made by counting those

remaining in the vial.

(Pincus 1958:1,335)

This strategy to ensure that women would adhere to the required test protocols did not work in all cases. In the publications of this trial Pincus reported seventeen pregnancies due to what he described as “patient failure”: these women had missed some days of tablet-taking (Pincus 1958:1,335; 1959:81).

Another, rather demanding requirement was that women had to submit themselves to regular physical examinations. Participants in the trial were expected to visit the Family Planning Association’s clinic, or an affiliated physician, for a pelvic examination including a biopsy of the endometrium of the uterus, a vaginal smear, and a series of blood samples. At home they were expected to collect 48-hour urine samples. These examinations and tests were aimed at monitoring any harmful effects of the progestins on important organ systems (Pincus 1958:1,335).

Disciplining women to submit themselves regularly to these examinations was not an easy task. For almost all the women the gynecological examinations were totally new. They were accustomed to seeing a doctor only when they gave birth (McLaughlin 1982:131). In the 1958 publication of the Rio Piedras trial, Pincus was able to report only on a relatively small number of women actually participating in the physical examinations: the blood samples were carried out in thirty-nine women, the urine tests in forty – two women, and the endometrial biopsies, obviously even harder to acquire, were obtained only from “some of these subjects” (Pincus 1958:1,338-1, 341).

The hardest quest was, however, to ensure that the women would continue to participate in the trial. At this point the field trial in Puerto Rico turned out to be rather problematic. The original object was to enroll 130 women, but for various reasons participants kept dropping out (Pincus 1958:1,335). In the Rio Piedras trial more than half of the women who started had withdrawn by the end of the year. Consequently, the trial did not meet Pincus and Rock’s expectations. It was quite clear that more data were needed, not only to convince their colleagues but, even more importantly, to serve as evidence toward approval of the new drug by the Food and Drug Association (FDA) (McLaughlin 1982:133).

Second and third trials were therefore quickly organized in Humucao, Puerto Rico, and a fourth one on another Caribbean island: Haiti (McLaughlin 19 8 2:133).36 These trials suffered from the same handicaps as the first trial, although they showed a less dramatic drop-out rate of approximately 10 to 30 per cent (Pincus 1959:8). Most women stopped the medication because of troublesome side effects they experienced during the pill-taking. Women participating in the trials had experienced unpleasant side-effects including dizziness, nausea and headaches, the most frequently mentioned complaints (Ramirez de Arellano and Seipp 1983:116). In the first reports of the trials Rice-Wray therefore concluded that the pill “caused too many side-reactions to be acceptable generally” (McLaughlin 1982:134). Pincus and Rock were also aware of possible adverse health effects of the hormones, particularly the growth of cervical, uterine or breast cancer. The participants of the trials therefore received regular physical examinations, including the vaginal smear test to screen for cervical cancer.

These tests revealed no risk in the incidence of uterine cancer…. On the basis of these early studies the American Cancer Society and the Andre and Bella Meyer Foundation made substantial grants to finance even more extensive examinations among women participating in the Puerto Rican and Haitian field trials, studies which have continued over a period of more than eight years.

(Maisel 1965:139)

In order to check for the complaints of nausea and headaches, Pincus designed a test in which a group of women were given placebos. From this trial Pincus concluded that such complaints had a psychosomatic origin (McLaughlin 1982:132). Other people involved in the trials attributed these reactions to the “emotional super-activity of Puerto Rican women” (Ramirez de Arellano and Seipp 1983:116). Pincus ascribed the side-effects “to the small amount of estrogen that occurred in progestin as a contaminant” and requested the Searle Company to synthesize a compound that would be free of estrogen. Subsequent trials with the pure compound, however, showed an increase in the incidence of break-through bleeding. Pincus and Rock ascribed this to the missing estrogen and therefore decided to include this hormone again in the contraceptive compound. This combined estrogen – progesterone preparation was given the trade name Enovid (Ramirez de Arellano and Seipp 19 83:115).37 Other reasons for dropping out were “moved or too distant, sterilized, separated from husband, disapproval of the husband, objections by the woman’s own doctor, protests of the woman’s priest” (Pincus 1958:1,344). The contraceptive pill had to compete with other contraceptive methods, in Puerto Rico particularly sterilization. Many women who dropped out of the clinical trials chose sterilization as “a more desirable solution to the problem of fertility control” (Ramirez de Arellano and Seipp 1983: 134). Keeping women in the trial turned out to be a hopeless task. In sum, after three years of trials all the original subjects had been replaced by other participants. Or, to quote Pincus: “We try to watch them as best as we can and we attempt to regain the withdrawals for our research. Actually, we reach approximately 20% ”.38

Pincus and Rock addressed the problem of drop-outs that could not be solved in the field in their reports. Here we see how scientific reports are not

simply a reflection of the proceedings of research, but are a far stronger tool than that: they are a representation that creates a new reality. Let us take a closer look at what happened in Pincus and Rock’s reports. In most of their publications the women participating in the trials have disappeared completely from the stage. They were replaced, quite simply, by the number of treated menstrual cycles. In the 1958 publication of the Rio Piedras trial, Pincus concluded: “In the 1279 cycles during which the regime of treatment was meticulously followed, there was not a single pregnancy” (Pincus 1958: 133). And the 1959 publication in Science, describing all four Caribbean field trials, reported:

We have recently collected and analyzed the data (to November 1958) from these four projects and present here the outstanding findings derived from these data; 830 subjects took the medication for a total of 8133 menstrual cycles, or 635 woman years.

(Pincus 1959:81)

A popular writer adopted this representation strategy as well. In The Hormone Quest the author concludes:

By 1960 1600 women at the Caribbean Field Trial Centers had used Enovid as a contraceptive for from a few months to nearly four years. Their experience, as a group, with the new steroid covered nearly 40, 000 menstrual cycles or—as medical statisticians prefer to put it—about 3000 woman-years of exposure to the possibility of pregnancy.

(Maisel 1965:134)

This representation strategy had important consequences. The very act of representing women as menstrual cycles resulted in a major increase of scale: the grand totals of the trials now included much more impressive numbers than a focus on the individual subject might have achieved. The trials were thus presented as having met their purpose: the testing of the progestins on large numbers of women over longer periods, as a prerequisite for its approval as a safe and reliable contraceptive.