Cartwright Inquiry


The Cartwright Inquiry was a committee of inquiry held in New Zealand from 1987 to 1988 that was commissioned by the Minister of Health, Michael Bassett, to investigate whether, as alleged in an article in Metro magazine, there had been a failure to treat patients adequately with cervical carcinoma in situ at National Women’s Hospital  by Herbert Green, a specialist obstetrician and gynaecologist and associate professor at the Postgraduate School of Obstetrics and Gynaecology, University of Auckland. The inquiry was headed by District Court Judge Silvia Cartwright, later High Court Justice, Dame and Governor-General of New Zealand. The Report of the Cervical Cancer Inquiry was released on 5 August 1988.

Background

A 1984 medical paper published in Obstetrics and Gynecology by colposcopist Bill McIndoe, pathologist Jock McLean and gynaecologist Ron Jones, who were all employed at NWH and colleagues of Green, as well as statistician Peter Mullins, described an audit of 948 women who had been diagnosed with cervical carcinoma in situ at New Zealand's National Women's Hospital from 1955 to 1976. The authors retrospectively divided the women with CIS into two groups: those with normal cytology follow-up at two years after initial management and those who continued to have abnormal cytology . Among those who continued to have abnormal cytology, a much higher proportion developed invasive cervical or vaginal vault cancer. The authors concluded that CIS of the cervix had significant invasive potential.
Using that paper, the medical experience of one of Green's patients as a case study and interviews with staff, a freelance journalist, Sandra Coney, and an academic, Phillida Bunkle, published an exposé, 'An "unfortunate experiment" at National Women’s Hospital', in Metro Magazine in June 1987. The article alleged that Green was carrying out research on his patients without their knowledge or consent and that since the mid-1960s, he had been withholding conventional treatment from some patients with CIS. Coney and Bunkle took the title of their article from a 1986 letter in the New Zealand Medical Journal by Professor David Skegg, a University of Otago cancer epidemiologist and expert in population screening in which he referred to "the unfortunate experiment at National Women's Hospital" in his reply to Green about the study. The authors claimed that no one could give them an assurance that the study had been stopped. Publication led to such enormous public outrage and concern that within ten days, the Minister of Health, Michael Bassett, established an inquiry.

Green's study

In June 1966, Green received permission from the NWH's Senior Medical Staff and the Hospital Medical Committee to undertake a study of following women with CIS without treatment. Green's proposal stated that all patients with a diagnosis of CIS under the age of 35, with positive smears and no colposcopic evidence of invasive cancer and without a ring or cone biopsy, would be followed. The Minutes recorded that Green stated that 'his aim was to attempt to prove that carcinoma-in-situ is not a pre-malignant disease'. He also said that if at any stage concern was felt for the safety of a patient, a cone biopsy would be performed.
Green published the evolving results of his study in the natural history of CIS in a series of papers from 1966 to 1975. In 1969, he wrote about his methods more clearly than in his proposal: "In 1965 the SMS of NWH initiated a project under the supervision of the author, for patients up to 35 years of age whose only abnormal finding was positive cervical cytology.... a histological diagnosis was to be established by punch biopsy of the most colposcopically significant area. Provided the biopsy did not remove the entire significant area, or reveal invasive carcinoma, there was to be no further treatment. As of the end of 1967, 33 such patients were studied." Green reported that none of the patients had been subsequently diagnosed with invasive cancer, as of June 1968.  
In 1970, Green published further results and described his study: "The only way to settle finally the problem of what happens to in situ cancer is to follow indefinitely patients with diagnosed but untreated lesions. This is being attempted." Again, he recorded no patients developing invasive cancer. A further paper from 1970 provided mathematical estimates of the invasive potential of CIS by using "follow-up observations on 75 patients who were untreated or incompletely treated". No patients were reported to have developed invasive cancer.
In the last paper, in 1974, Green again described his study: "This series of 750 cases of in situ cervical cancer, and the following of 96 of them with positive cytology for at least two years, represents the nearest approach yet to the classical method of deciding such an issue as the change of one disease state to another – the randomised controlled trial. It has not been randomised and it is not well controlled, but it has at least been prospective." This time, Green reported that 10 cases had apparently progressed to invasion but that only 2 had "no clinical or histologic doubt." He concluded that the present series "offers little proof of the progression hypothesis."
Green published nothing else. A later and separate re-evaluation of the invasive potential of CIS, based on a review of NWH patients' files, was published by McIndoe et al, with very different findings.

Inquiry

District Court Judge Silvia Cartwright was appointed by warrant dated 10 June 1987 as a Committee of Inquiry to inquire into the treatment of cervical cancer at the National Women's Hospital and other matters.

Terms of reference

The terms of reference for the inquiry were to investigate :
  1. Whether there was a failure adequately to treat cervical carcinoma in situ at the NWH, and if so, the reasons for that failure and the period for which that failure existed;
  2. Whether a research programme into the natural history of CIS of the genital tract was conducted at NWH and, if so:
  3. # Whether it was approved before being instituted or while it was underway;
  4. # Whether patients were aware at any stage that they were participants in a research programme;
  5. # Whether any expressions of concern about the programme were considered and investigated at the time and if so, by whom.
  6. Whether there was a need to contact women involved with a view to providing further advice or treatment to them;
  7. Whether the National Women's Hospital's procedures for approval of research and/or treatment and its surveillance were adequate, in particular whether they ensure that the rights of patients are protected;
  8. What steps, if any, need to be taken to improve the protection of patients in respect of whom research and/or treatment is conducted;
  9. Whether the patients at the National Women's Hospital were properly informed of the treatment and options available to them, and, if not, the steps needed to see that they are;
  10. The training given to medical students and medical practitioners regarding the proper detection and treatment of cervical cancer and pre-cancerous conditions of the genital tract, and whether steps should be taken to improve this training or to inform previous trainees;
  11. To inquire into the relationship between the academic and clinical units at the Hospital
  12. Any other matter which is relevant.

    Procedure of the inquiry

The first preliminary hearing was held on 18 June 1987. Three medical advisers were appointed to assist and advise Judge Cartwright, Professor E V MacKay, Professor of Obstetrics and Gynaecology at the University of Queensland, Dr Charlotte Paul, Epidemiologist, University of Otago Medical School and Dr Linda Holloway, Pathologist, University of Otago Wellington School of Medicine. Counsel assisting the inquiry were Lowell Goddard QC and Philippa Cunningham.
Public hearings commenced on 3 August 1987 and ran for seven months. Evidence was heard from 59 witnesses, including from medical experts brought from around New Zealand, as well as from Australia, United States, Japan, UK, and Norway. Twelve patients or former patients, including ‘Ruth’, and two relatives of patients gave evidence in the absence of the public and the media. Judge Cartwright heard evidence privately from a further 70 patients. The medical advisers carried out a review and analysis of patient files. Overall 1200 patient files were reviewed.

Report findings in respect of Green's study

In respect of Term of Reference One, the Report defined ‘adequate treatment’ of CIS since the 1950s as being that based on generally accepted treatment together with evidence of eradication of the disease, and concluded that, by that standard, there had been a failure to adequately treat a number of women with CIS at NWH, For a minority of women, their management resulted in persisting disease, the development of invasive cancer, and in some cases, death. The reasons for the failure were the implementation of the ‘1966 trial’, failure to recognise the dangers for patients, failure to evaluate the risks to patients if Green’s hypothesis was wrong, failure to note the rising incidence of invasive cancer, to stop the trial and treat the patients as soon as cogent evidence of risk began to emerge, and failure on behalf of some colleagues and the NWH administration to impinge on Green’s clinical freedom and act decisively to stop the trial in the interests of patient safety.
In respect of TR2, the Report found that the 1966 trial was a research study into the natural history of CIS. The 1966 trial was approved by the NWH Hospital Medical Committee on 20 June 1966.The great majority of patients did not know, except intuitively, that they were participants in the trial. ‘Had patients been informed of their inclusion in the trial, informed of the types of treatment available to them, informed of the risks of procedures that were not conventional, definitive treatment for CIS, and given the opportunity to decide whether or not to be part of the trial, then the trial could not be so severely criticised’.
Green's published work showed evidence of scientific fraud, according to Judge Cartwright. "An analysis of Dr Green's papers point to misinterpretation or misunderstanding of some data on his part, and on occasions, manipulation of his own data." "The inference to be drawn from Dr Green’s proposal and published papers is that CIS will progress to invasive cancer in only a very small proportion of cases if at all. This inference is incorrect and reliance on it has been dangerous to patients."
The New Zealand Medical Council accepted that the 1978 protocols laid down by the Hospital Medical Committee spelled the end of the 'trial'. The International Federation of Gynecology and Obstetrics published statistics from the late 1970s which showed that the hospital ranked among the best institutions in the world in regard to five-year survival rates for patients with carcinoma of the cervix.
For the most comprehensive analysis of the Cartwright Report and its relevance to modern screening programmes, see the 2019 encyclopaedic guide to screening published by Oxford University Press which draws on the Inquiry as a case study.