Professor Linda Bryder1 argues that the decades of limited or withheld treatment of CIS (CIN3) of the cervix at National Women’s Hospital which ended with the Cartwright Inquiry of 1988 has been retrospectively justified by recent new research which once again blurs the boundary between CIN 2 and CIN 3. I argue, however, that Bryder’s criticism of the Cartwright Inquiry misses the key issue. The inquiry found evidence of significant scientific misconduct in the research which justified the hospital’s limited approach.
The primary finding investigated by the Cartwright Inquiry was that information generated by Associate Professor Green and propagated by National Women’s Hospital about CIS of the cervix was, at best, erroneous and much of the treatment based on it was harmful.
The focus should not, therefore, be on Green as an individual. The impropriety involved in the “Unfortunate Experiment” as outlined in Dame Sylvia Cartwright’s Report was systematic scientific misconduct.2 It is the duty of research institutions to uphold scientific principles and ethical standards and to enforce rigorous scientific process. The outstanding lesson of the “Unfortunate Experiment” was the failure of the institutions (which employed and protected Green) to fulfil their professional, academic and intellectual responsibilities.
Green maintained that CIS and invasive cervical cancer were separate conditions; one did not progress to the other. A single instance of progression should have been enough to invalidate his hypothesis and end the investigation. Instead, when examples of progression occurred, Green was allowed to change some and ignore other cases. It is worth recalling how a number of scientific principles were violated: attempted tampering with the labels on the smears and slides; retrospectively changing some diagnoses, claiming that they had been invasive all along (the so called “colposcopic misses”); reclassifying cases; and publishing these unscientific, erroneous results. In doing so, Green delayed and compromised treatment—sometimes fatally—and he and his team and the responsible institution failed to inform the women at risk.
Hospital specialists McIndoe and McLean, who recorded and analysed Green’s activities, accumulated case notes of an increasing number of cases of invasion and documented the consequences of extended and extreme treatment—and even death—which resulted.
They submitted their findings to those in positions of responsibility for management at the hospital and the University of Auckland. Hospital management avoided the issue and the university was slow to react, putting more women at risk for longer. Appropriate scientific principles to investigate and analyse these cases were not used: the selected time frame was arbitrary; no inclusion/exclusion criteria were established or applied to the review; only a part of the available evidence was examined; and the significance of the remaining cases were minimised. There was no action resulting from either the clinical or scientific issues raised. Rather, those who raised them paid a heavy professional price and were marginalised within the institution.3,4
The judicial nature of the Cartwright Inquiry allowed access to the original primary research data for examination. Its conclusions rested upon analysis of these data. In stark contrast, Bryder, unlike some scientists,5 did not access or examine the original case files on which the inquiry findings depended and she is therefore not in a well-informed position to rigorously evaluate either the evidence presented to the inquiry or the conclusions which flowed from it.
The Cartwright Inquiry showed that independent oversight can safeguard the public interest. It is a safeguard that may be needed in the future. Most scientific research data are today generated by commercial interests or under contract and protected from disclosure.
Commercialised, corporatised science is thus responsive to the market but impervious to the voice of concerned or damaged consumers. One effect of Bryder’s continued condemnation of the inquiry is to invalidate much needed processes of independent, external evaluation.
In addition, critics of Bryder’s work have found academic avenues for open debate and questioning inaccessible.6 Good Science arises from reasoned critique. It sometimes takes a judicial process to police Bad Science and in this the Cartwright Inquiry was important. The problem is not just that patients are not told, it is that what they are told may not be scientifically valid. Valid information is a prior, necessary condition for informed consent.
The strength of the Cartwright Inquiry was that it showed, on careful evaluation of the evidence, that patients were given inadequate information. It set out to analyse why this had occurred and it recommended changes to ensure that it would never happen again. In this way it provided a touchstone for the better in the relationship between patients and medical staff in New Zealand and drew attention to primum non nocere: first do no harm.