Silverwood, Monica --- "Is the Non-Consensual Use of Women’s Health care Records to Evaluate the National Cervi cal Screening Programme Justified?" [2006] NZLawStuJl 3; (2006) 1 NZLSJ 1

Last Updated: 24 October 2012

IS THE NON-CONSENSUAL USE

OF WOMEN’S HEALTHCARE RECORDS TO EVALUATE

THE NATIONAL CERVICAL SCREENING PROGRAMME JUSTIFIED?

_{MONICA SILVERWOOD}^∗

Introduction

Cervical cancer is one of the few cancers that can be prevented. A screening programme that tests women on a regular basis can detect pre-cancerous lesions and treat them before invasive cancer develops.¹

To be of real benefit to women, the screening programme itself also needs to be regularly monitored. An important part of that monitoring is investigation of the smear histories of women enrolled in the programme and those diagnosed with cervical cancer to determine why cancers still develop despite the existence of a screening programme. An evaluation of this kind uses information that identifies individual women and also information beyond that which is held on the programme’s register, including primary health care records. Until very recently, evaluators of New Zealand’s National Cervical Screening Programme (the NCSP) had to obtain the consent of individual women to access this kind of personal identifying health information.² This requirement meant that a comprehensive and statistically meaningful evaluation of the NCSP’s performance in its first decade of operation was never completed.³ To ensure effective evaluation can

^∗ LLB (Hons) BA, University of Canterbury.

¹ Kirsten Coppell, Charlotte Paul and Brian Cox, ‘An Evaluation of the National Cervical

Screening Programme Otago Site’ (2000) 113(1104) The New Zealand Medical Journal 48, p.

^48.

² Section 74A(5) of the Health Act 1956 prevented disclosure of information on the

^{NCSP Register without the woman’s consent.}

³ Euphemia McGoogan, Report on the National Cervical Screening Programme and progress towards Implementation of the Gisborne Inquiry Recommendations (June 2003) http://www.csi.org.nz/other_reports/CSIMcGooganFINALREPORT.pdf [20]. Also see the Gisborne Cervical Screening Inquiry, Report of the Ministerial Inquiry into the Under- reporting of Cervical Smear Abnormalities in the Gisborne Region (2001) http://www.csi.org.nz/report/table_of_contents.htm, [9.5].

2 The New Zealand Law Students’ Journal (2006) 1 NZLSJ

be carried out, the Health Act 1956 was amended in March 2004 to allow evaluators full access to a woman’s health information held by the NSCP, the Cancer Registry and any health practitioner, laboratory or hospital, without her additional consent.⁴

The Amendment Act was controversial as many women were concerned about evaluators having access to sensitive and intimate information about them, without their knowledge or consent. The legislation is based on the decision that this invasion of privacy is justified by the need to properly evaluate the NCSP and the resulting benefits of that which will accrue to all women. It is an example of the classic public health dilemma: the clash of individual and community interests. The public controversy and unease surrounding this aspect of the Amendment Act make it worthy of critical attention. This essay will examine the justifications for removing the requirement of individual consent to access primary and hospital care records for evaluation of the NCSP. They will be assessed against the standard that, especially in the health care context, any such incursion into personal autonomy ought to be only what is necessary and as minimal a derogation as possible. On the whole, the policy does conform to these parameters however there are aspects that remain a cause for concern for women. The apprehension about these changes also provides an insight into how New Zealanders view the balance between confidentiality and public health. The power to adopt this policy is not being questioned; the focus is on whether the justifications given for it are sufficient to quell public concern.

A: The National Cervical Screening Programme

The National Cervical Screening Programme was established in 1990 in response to a recommendation in the 1988 report of the Cartwright Inquiry.⁵ The committee, led by Judge Silvia Cartwright (as she then was), investigated a long-running research study at National Women’s Hospital in Auckland where women underwent repeated cervical smear tests and biopsies without being offered adequate treatment for

⁴ Health (National Cervical Screening Programme) Amendment Act 2004.

^{5 Committee of Inquiry into Allegations Concerning the Treatment of Cervical Cancer at}

National Women’s Hospital and into Other Related Matters, Report (The Cartwright Report)

^{(1988). Also see Gisborne Cervical Screening Inquiry, Report, supra n. 3, [3.2].}

Non-Consensual Use of Healthcare Records 3

abnormalities that were discovered.⁶ The majority of women were not informed that they were participating in a research study and it appeared that obvious symptoms of invasive cancer were overlooked or downplayed. Some health professionals had been calling for New Zealand to follow the lead of other countries and establish a cervical screening programme and the Cartwright Inquiry recommendations settled the matter.⁷ It is important to recognise that New Zealand’s cervical screening programme was at its inception, and continues to be, closely tied to the sad events at National Women’s Hospital. Rightly or wrongly, this has certainly influenced the way many women feel about the Amendment Act’s provisions.⁸

The NCSP is an ‘opt-off’ programme: women are automatically enrolled after their first smear test or other relevant procedure but they can choose to withdraw at any time. This ensures the Programme has a high participation rate which is essential to its success in reducing incidence of, and mortality from, cervical cancer. Women must be notified of their enrolment and at that time provided with information about regular screening, the NCSP, what information is collected and held, who has access to it and what for.⁹ The NCSP Register holds the results of a woman’s smear tests, her contact details and other identifying information, including her ethnicity. The core functions of the screening programme are to ensure that women are tested regularly (every three years in New Zealand) and that tests that report abnormalities are followed up appropriately. To this end the NCSP uses the information it holds to remind women when they are overdue for a smear test and to contact women who need further treatment.

Since becoming operational in 1990, the NCSP has contributed to a nearly 40 percent reduction in the incidence of cervical cancer and a 46

⁶ The study ran from 1966 to the mid 1980s. The practitioners involved were testing the theory that carcinoma in situ was not a precursor to invasive cervical cancer.

^{7 Gisborne Cervical Screening Inquiry, Report, supra n. 3, [3.2].}

⁸ McGoogan [Recommendation 11.13]. Professor McGoogan notes in that paragraph that “[t]he time has arrived for the National Screening Programme to be treated as a medical programme which is part of a national cancer control strategy. In the past its link with the Cartwright Report has at times resulted in its purpose as a cancer control strategy being compromised for non-medical reasons.”

^{9 Health Act 1956, section 112F.}

percent reduction in deaths from the disease.¹⁰ Even so, there have been failings. The most notable is of course the under-reporting of abnormalities in cervical smear slides in the Gisborne region which led to the Gisborne Cervical Screening Inquiry in 2000. The Inquiry Report released in 2001 revealed many specific failings with regard to the practice of cervical cytology in certain laboratories but also identified systemic issues that contributed to the problems of the NCSP. One of these was the “failure to ensure that there was legal power to do what was needed for the Programme to be effective.”¹¹

B: The Health (National Cervical Screening Programme) Amendment Act 2004

The Gisborne Inquiry Report described the law surrounding evaluation of the NCSP as a “legal quagmire.”¹² In order to determine whether there had been systemic under-reporting on a national scale¹³ and whether invasive cervical cancers were detected by regular screening, an evaluation of the smear histories of individual women had to be carried out. Achieving this would mean accessing identifiable information on the NCSP Register. Section 74A(5) of the Health Act however effectively prevented any such evaluation taking place.¹⁴ It stated that “no person may disclose information on the register that identifies a woman” without her consent, and none of the listed exceptions applied to the independent auditors. To overcome this, the Inquiry Committee issued a subpoena under section 4D of the Commissions of Inquiry

¹⁰ Hazel Lewis, ‘New cervical screening legislation: access to clinical records’ (2004) 117 (1191) The New Zealand Medical Journal 829, p. 829.

¹¹ Gisborne Cervical Screening Inquiry, Report, supra n. 3, [6.74].

^{12 Ibid.}

¹³ Term of Reference Three for the Gisborne Cervical Screening Inquiry, supra n. 3.

^{14 The Gisborne Inquiry also outlined other legal barriers that prevented the evaluation}

taking place. The Privacy Act 1993 regime (specifically Rule 11(2)(c)(iii) of the Health Information Privacy Code) was mistakenly applied to the identifiable information held on the national Cancer Registry which required the evaluators to obtain ethics committee approval for disclosure of the information. The ethics committees around the country however held that the evaluators had to obtain consent before accessing the identifiable information. This put evaluators in an impossible situation, unable to access any information that would identify women who had developed cervical cancer. See Gisborne Cervical Screening Inquiry, Report, supra n. 3, [6.89].

Act 1908¹⁵ requiring the Ministry of Health to produce the relevant information to the evaluators. The Ministry resisted production under section 74A(5) of the Health Act. The impasse could not be broken and the Committee considered referring the question of its subpoena power to the High Court for a ruling.¹⁶ This was abandoned however as a draft of the New Zealand Public Health and Disability Act submitted to the Committee removed this power from Committees of Inquiry into health matters, making the question purely academic in the view of the Committee members.¹⁷ The concern of the Committee that an audit of cervical cancer cases be carried out was also somewhat allayed by Ministry of Health assurances that their national evaluation of the NCSP would include a cancer audit.¹⁸

Even if the Ministry had released the information on subpoena however, the resulting legal situation would hardly have been desirable as evaluation would require the relevant information to be produced by compulsion. In addition, had the evaluators been able to progress to an audit of individual cases of cervical cancer as they intended, the consent of each woman would need to be obtained to access her primary and other health care records. This too was a less than ideal situation for the evaluators, a point that will be discussed below.

The Ministry had failed to ensure the necessary legal powers to administer the NCSP were present, and the recommendations of the Gisborne Inquiry Report were unequivocal: section 74A of the Health Act 1956 had to be amended to give appropriately qualified evaluators ready access to all information on the NCSP¹⁹ and all medical files of women diagnosed with cervical cancer recording the treatment of the cancer by all health providers involved.²⁰ This is precisely what the Amendment Act has achieved.

¹⁵ The Gisborne Inquiry Committee was established under section 47 of the Health and Disability Act 1993. Section 47(3) of that Act gave the Inquiry Committee the powers of a Commission of Inquiry under the Commissions of Inquiry Act 1908.

^{16 Commissions of Inquiry Act 1908, section 10.}

¹⁷ The relevant provision is section 77(j) of the New Zealand Public Health and Disability Act 2000. Also see Joanna Manning, ‘Review: Health Care Law’ [2001] New Zealand Law Review 189, pp. 192-193.

^{18 Gisborne Cervical Screening Inquiry, Report, supra n. 3, [31].}

¹⁹ Ibid., [11.14].

^{20 Ibid., [11.17].}

The purpose of the cervical screening programme is to reduce the incidence and mortality rates of cervical cancer. The Amendment Act supports this by assisting the evaluation of the NCSP through easier access to a wider scope of information. One of the objectives of the NCSP listed in the Amendment Act is to facilitate ‘continuous quality improvement’ by allowing and performing regular evaluations of the Programme.²¹ Evaluations are specifically stated to include the review of and investigation into the cases of individual women.²² In accordance with this new emphasis on evaluation, the Act provides for screening programme evaluators to have full access to relevant information about individual women held by the NCSP, the Cancer Registry, and records or specimens held by hospitals, laboratories and any health practitioner, including primary care records.²³ Additional consent from the relevant women is not required and health practitioners must comply with an evaluators’ request.²⁴ ‘Relevant women’ are those enrolled on the NCSP (whether or not they develop cervical cancer) and women who develop cervical cancer (whether or not they are enrolled on the NCSP).²⁵ In contrast to its predecessor, the Amendment Act gives evaluators fairly extensive powers of access. There are important limits however. The information evaluators collect must be relevant to their evaluation²⁶ and the Director-General of Health can impose specific limits on what information can be accessed.²⁷ There are also strict confidentiality requirements placed on the evaluators,²⁸ similar to the duty imposed on a medical professional in a therapeutic relationship.

C: The competing principles

It is the ability of screening programme evaluators to access the primary care and other health records of women without their specific knowledge or consent that has caused the most controversy. Green Party MP, Sue Kedgley, described the law change as “tragic” and stated

²¹ Section 112A

^{22 Section 112T}

²³ Section 112X

^{24 Section 112ZB}

²⁵ Section 112X(1)

^{26 Section 112X(4)}

²⁷ Section 112U(3)

^{28 Section 112Y(3)}

that it “spells the end of doctor-patient confidentiality for the one million women enrolled on the NCSP.”²⁹ On this view, women are forced to choose between confidentiality of their medical records and their health. Other writers have also been sceptical of the Amendment Act and questioned the efficacy of the limits placed on evaluators.³⁰

Many women are concerned that evaluators will have access to sensitive personal information, specifically about intimate matters such as terminations, sexual abuse, sexually transmitted infections, and psychiatric conditions.³¹ The concern is that much of this information could be considered relevant to an evaluation of cervical cancer as most cases of the cancer are caused by the sexually transmitted, and very common, human papilloma virus.³²

The 2004 Health Amendment Act exemplifies the classic public health dilemma where personal autonomy and privacy is in opposition with the public good. The confidentiality of health records “is a subject in which autonomy and beneficence clash head on.”³³

The individual’s right to privacy is “essential to human dignity and liberty”³⁴ and is recognised as such in many jurisdictions, often at the constitutional level. In terms of medical ethics it is an aspect of consumer autonomy and in this specific context it is manifested in the right to have health information kept confidential. People have a

²⁹ Sue Kedgley, A million NZ women lose their medical privacy tomorrow, Press Release (6 March

^{2005), http://www.greens.org.nz/searchdocs/PR8368.html. The New Zealand Medical}

Association also expressed concern about the harm that might be done to the doctor/patient relationship. See Sandy Elkin, ‘Bioethics Commentary’ [2004] February New Zealand Bioethics Journal 4, 5.

^{30 Elspeth McLean, ‘Dear Aunty Karen’ Otago Daily Times (Dunedin), 22 March 2005, 15}

and

^{Elspeth McLean, ‘Prayer for the screening programme’ Otago Daily Times (Dunedin), 14}

April 2004, 25.

^{31 Joanna Manning, ‘Health Care Law Part 2 – Legislative Developments’ [2004] New}

Zealand Law Review 385, 399.

^{32 About 70 percent of cervical cancers are caused by two different strains of this virus.}

See Denise Grady, ‘Vaccine Prevents Most Cervical Cancer’ New York Time (New York, USA), 7 October 2005, http://www.nytimes.com (Copy on file with author).

³³ Sir Douglas Black, ‘Access to records for epidemiological research’ (1992) 304 British Medical Journal 987, quoted in Professor David Skegg’s Second Brief of Evidence at the Gisborne Cervical Cancer Inquiry (2001) . http://www.csi.org.nz/proceedings/skegg2nd.pdf, para. [32].

³⁴ Barbara von Tigerstrom et al, ‘Legal Regulation of Cancer Surveillance: Canadian and

^{International Perspectives’ (2000) 8 Health Law Journal 1, p. 16.}

discrete autonomy interest in controlling access to their personal health information, regardless of consequence.³⁵ In New Zealand this interest is protected by the Health Information Privacy Code (the HIPC). The basic rules relevant here are first, that personal (identifiable) information can only be used for the purpose for which it was collected³⁶ and secondly, the health agency cannot disclose the information unless one of the listed exceptions applies such as when the individual consents or where consent cannot be obtained and the disclosure is necessary to prevent or lessen a serious and imminent threat to public health or safety.³⁷ The HIPC clearly covers primary care records that are the subject of concern here. Preserving the confidentiality of this information is important not only for consumer autonomy but also to protect the therapeutic relationship: a patient must be able to share the intimate, and sometimes embarrassing, details of their health to receive the best care. Following the passage of the Amendment Act, some writers expressed the concern that women would opt off the NCSP to prevent this invasion of their privacy.³⁸

The other principle in contention in this debate is the good of public health. While there is no specific legal duty on governments to advance public health it is certainly a moral imperative and a strong incentive to override the individual’s right in confidentiality of their information. Historically, public health activities have “no doubt exerted much more of a salutary effect on the longevity and physical, mental, and social well-being of the general population...than have advances in clinical medicine, important though the latter may be.”³⁹ The ethical principle of beneficence is clearly applicable here. In the context of cervical screening programmes it is essential that smear tests are read accurately so that abnormalities are detected early and treated. To ensure this is happening, the programme must be evaluated and monitored. A screening programme operating without evaluation puts women at a

³⁵ Helen Davidson, John Dawson and Andrew Moore, ‘Law, Ethics, and Epidemiology: The Case of the Cervical Screening Audit’ (2001) 2 (2) June New Zealand Bioethics Journal 8, p. 20.

^{36 Rule 10.}

³⁷ Rule 11.

^{38 Kedgley, Press Release, supra n. 29 and Manning [2004], supra n. 31, p. 400.}

³⁹ Marshall B. Kapp, ‘Book review Essays’, reviewing Lawrence O. Gostin, Public Health

^{Law: Power, Duty, Restraint (2000), (2001) 22 Journal of Legal Medicine 581, p. 581.}

greater risk by giving them, and their doctors, false assurance.⁴⁰ It is unethical to “exhort apparently healthy people to undergo medical procedures, when adequate steps cannot be taken to monitor the quality of the process or the outcomes achieved.”⁴¹ There is an obligation to evaluate.

There is a risk of very serious harm to the community in allowing personal interests in confidentiality to prevent the proper evaluation of the NCSP, as evidenced by the Gisborne Inquiry. When the conduct of an individual can cause harm to the community, the question of whether the general welfare will be promoted by interfering with the individual’s right becomes open to discussion.⁴² With regard to evaluation of the NCSP, it was decided that the general welfare would indeed be significantly promoted by removing the need for additional and specific consent from women to access their health records, overriding the usual privacy protections. The law certainly allows for this resolution of the debate in favour of the public health interest. The right to privacy, though fundamental, is not absolute. The exceptions to the rule of non-disclosure in the HIPC show how even in the context of sensitive health information, the right to privacy and confidentiality can be limited. More importantly, the rules in the Privacy Act and HIPC give way to any contrary statutory provision that authorises or requires the disclosure of information.⁴³

Before considering the justification for the Amendment Act policy it is necessary to address an exception to the rule of non-disclosure that could be relevant here. Rule 11(2)(h)(ii) provides that a health agency may disclose information if they believe on reasonable grounds that it is either not desirable or not practicable to obtain authorisation from the individual for the release, and the disclosure is required for a professionally required external quality assurance programme. Evaluators cannot however rely on this provision to gain access to the information they need. First, the health agency may release the

⁴⁰ Gisborne Cervical Screening Inquiry, Report, supra n. 3, [6.72].

^{41 Professor David Skegg quoted in Gisborne Cervical Screening Inquiry, Report, supra n.}

3, [6.72]. See also Davidson, Dawson and Moore, supra n. 35, p. 13.

^{42 Margaret Brazier and John Harris, ‘Public Health and Private Lives’ (1996) 4 Medical}

Law Review 171, p. 177, quoting John Stuart Mill, On Liberty (1869).

^{43 Privacy Act 1993 section 7.}

information; the power is discretionary.⁴⁴ Secondly, the usual practice regarding external evaluations is to require ethics committee approval and as was noted above, this cannot be guaranteed. The Amendment Act is rightly characterised as a statutory override of the HIPC rules rather than falling under an exception.

D: Justifying the removal of the consent requirement

The subpoena issued by the Gisborne Inquiry Committee, their recommendations for legislative change and the resulting Health Amendment Act all rely on the same core rationale to override the individual interest in confidentiality of medical records: that it is necessary to ensure the quality and effectiveness of what is a very valuable screening programme and that the legislation only goes so far as is necessary to meet this need.

1. Establishing the need to override

The necessity argument is grounded in the value of the cervical screening programme to women. Both internationally and in New Zealand screening programmes have been shown to significantly reduce the incidence of, and number of deaths from, cervical cancer.⁴⁵

That the NCSP is a good and valuable programme, and that it needs to be monitored to remain so, cannot be disputed. The facts that gave rise to the Gisborne Inquiry make it plain that a great deal of harm can result from poor, or in fact non-existent, quality control. The ethical principle of beneficence is very important in this context as it is in the best interests of all women that the cervical screening programme is of a high standard. The value of the screening programme can only be maintained through evaluation.

Attention must then turn to the requirements of such an evaluation. The aim is to assess the appropriateness of follow up and treatment of women with abnormal smears and to determine why cases of cervical

⁴⁴ Dr Tricia Briscoe notes, and it is discussed below, that in some instances medical practitioners would be ethically bound to refuse evaluators access to patient files. See Dr Tricia Briscoe, ‘Medical Ethics – Curly Questions’ (Paper presented at the Fifth Medical Law Conference, Wellington, 30 May 2004). http://www.nzma.org.nz/news/issues/medical-ethics.pdf (Copy on file with author)

^{45 Coppell, Paul and Cox, supra n. 1, p. 48.}

cancer still occur despite the existence of the screening programme.⁴⁶

This type of evaluation requires access to the smear test results and other relevant medical histories of individual, identifiable women. A vital element of any meaningful and effective evaluation is a high participation rate among the population being assessed – accurate results depend on complete data.⁴⁷ Without this, the study will yield biased conclusions that cannot be relied upon and the evaluation is pointless.⁴⁸ Further, as the number of women whose histories need to be reviewed is small, access to close to one hundred percent of records is needed for the results to be epidemiologically meaningful.⁴⁹

Researchers proposing evaluations of the NCSP have argued that it is not possible to obtain this level of consent.⁵⁰ This was exactly the case for the Ministry of Health review of the NCSP conducted in 2001, although it must be noted that a 2004 Auckland University audit of cervical cancer cases suggests the difficulties of securing the consent of women are not insurmountable.⁵¹ Even so, 19.2 percent of the women contacted for the Ministry’s review refused access to their records and

24.3 percent of women could not be located. The overall response rate was only 56 percent and it was even lower for Mäori (28 percent) and Pacific Island (31 percent) women who have the highest incidence of, and mortality rates from, cervical cancer.⁵² These figures support Parliament’s conclusion that it is simply not desirable or practicable to require women’s consent. The provisions of the Amendment Act and the recommendations of the Gisborne Inquiry Committee are clearly grounded in the belief that effective evaluation cannot be guaranteed if

⁴⁶ Lewis, supra n. 10, p. 829.

^{47 Davidson, Dawson and Moore, supra n. 35, pp. 9 and 14.}

⁴⁸ Ministry of Health, Progress in Implementing the Recommendations of the Cervical Screening Inquiry (February 2002), http://www.oag.govt.nz/HomePageFolders/Publications/Cervical_Screening/cervical- screening.pdf, [5.10].

⁴⁹ Lewis, supra n. 10, p. 829. Access to clinical records will focus on the group of about

^{150 to 200 women who develop cervical cancer each year.}

⁵⁰ Elkin, supra n. 29, 5. See also McGoogan, supra n. 3, p. 7.

^{51 University of Auckland, The New Zealand Cervical Cancer Audit Report 2000 – 2002, (2004)}

http://www.moh.govt.nz/cervicalcanceraudit.

^{52 Ministry of Health, Progress in Implementing the Recommendations, supra n. 48, [5.3 n. 22].}

Regarding cervical cancer incidence in Mäori women see University of Auckland, The

^{New Zealand Cervical Cancer Audit, supra n. 51, p. 1.}

women’s consent is required. From this view, it is a stark choice between privacy and safety.⁵³

2. A limited override

Supporters of the Amendment Act provisions emphasise that evaluators will only be concerned with women’s smear histories and that they will be bound by stringent confidentiality requirements.⁵⁴

^{Health practitioners can also oversee an evaluator’s access to their}

patient files.⁵⁵ These limits are designed to give women greater confidence in the integrity of the NCSP and assurance that there is no threat to the confidentiality of the doctor-patient relationship or to a woman’s privacy. This is because the powers are (arguably) limited to what is strictly necessary to complete an effective evaluation.

The Gisborne Inquiry and independent experts have strongly argued that evaluator access to health care records cannot be considered a dangerous foray into personal privacy as evaluation of care is a ‘part in parcel’ of treatment.⁵⁶ The proposed audit was described as “a routine component of quality control in many countries.”⁵⁷ An example of this is Scotland, where there is a public health audit of the women who develop cervical cancer.⁵⁸ The point is that evaluation of the screening programme is an integral aspect of the treatment women receive through the programme, not an external or additional imposition on them.

A further protection for women is that the Director-General of Health is obligated to inform them that evaluation is one reason their information is collected and what exactly evaluators will have access to and how they will use it.⁵⁹ The argument is that any perceived intrusion in privacy is therefore lessened and no longer represents a harm to

⁵³ C Paul, ‘Should the law require doctors to make records available for audit of cervical screening?’ (2001) 114 (1143) The New Zealand Medical Journal 499, p. 499.

^{54 Lewis, supra n. 10, p. 829 and Health Act 1956, sections 112X(4), 112U(3) and 112Y(3)}

⁵⁵ Health Act 1956, section 112X(6).

^{56 Gisborne Cervical Screening Inquiry, Transcript, 25 September 2000}

http://www.csi.org.nz/proceedings/CSIev25092000.pdf, C/601, lines 1-6.

^{57 Skegg, Second Brief of Evidence, supra n. 33, [19].}

⁵⁸ Ibid.

^{59 Health Act 1956, section 112F.}

women. In essence, they know about the evaluation and so are consenting to the use of their information for that purpose. The final factor to consider is that most women, when approached, would consent to evaluators having access to their medical records for quality assurance purposes.⁶⁰ Essentially, the policy makers have arrived at the same decision the majority of women would themselves make, given the chance. Despite the inconsistency with the concerns about low consent rates noted above, respect for hypothetical consent does appear to be an underlying justification for the Amendment Act provisions. The Gisborne Report stated that “[i]t is difficult to see why women might object to an independent evaluation team seeing information to which those medical persons who are involved in their treatment have unrestricted access.”⁶¹

These arguments work to justify overriding confidentiality not by establishing a need to do so, but by demonstrating that there is either no real invasion of privacy at all, or that it is such a minimal incursion as to not attract any significant weight when considering the individual/community balance.

E: Assessing the legitimacy of justification arguments

The personal autonomy of women has assumed a position of great importance in relation to New Zealand’s cervical screening programme. For many people, it remains linked to the abuses of trust at Auckland National Women’s Hospital that led to the Cartwright Inquiry. Overseas reviewers of both the NCSP and BreastScreen Aotearoa have recognised that this has led to an over zealous protection of patient confidentiality and informed consent and a strong ‘privacy at all costs’ mentality among the public and health professionals. ⁶² There is certainly truth in this statement, particularly evident in the now repealed section 74A(5) of the Health Act. The Amendment Act 2004 withdraws this promise of confidentiality⁶³ in favour of the public health need to evaluate but the significance of individual women’s

⁶⁰ Davidson, Dawson and Moore, supra n. 35, pp. 20-21.

^{61 Gisborne Cervical Screening Inquiry, Report, supra n. 3, [6.98].}

⁶² McGoogan, supra n. 3, [43]. Professor McGoogan refers to the independent review of

^{BreastScreen Aotearoa conducted in February 2002 by Professor Jocelyn Chamberlain.}

⁶³ Privacy Commissioner (NZ), Improving the National Cervical Screening Programme (23 July

^{2001) http://www.privacy.org.nz/people/cervscrn.html.}

autonomy cannot be relegated completely. If it is accepted that confidentiality of medical records is an important right, then a fair distribution of benefit and burden will be when that right is infringed to the smallest extent necessary in order to meet the public health need.⁶⁴

On a broad analysis the policy of the Amendment Act 2004 does meet the least harm standard. First, there is a pressing need to evaluate the NCSP and audit cases of cervical cancer. Secondly, overriding the consent requirement is the best way to guarantee a meaningful evaluation and the access powers of evaluators are constrained. On a closer inspection however there are a number of issues that give legitimate cause for concern.

1. Opting off the NCSP

A decision to opt off the NCSP is, in many respects, an empty gesture as it does not ensure the privacy of a patient’s records. ⁶⁵ The smear history and treatment record of a woman who opted off the NCSP and is diagnosed with cervical cancer can provide valuable information about the screening programme so the records of this class of women have been brought within the scope of an evaluator’s access powers. Their epidemiological value however, does not seem to provide sufficient justification for accessing the records without the individual woman’s knowledge or consent. In such a case, the Amendment Act is not merely removing the need for a positive consent; it is overriding a decision that is analogous to a competent refusal of medical treatment. As stated by the New Zealand Law Society in their submission on the Bill, “[i]t seems hard to justify the non-consensual accessing of a person’s medical records to evaluate a screening programme in which that person does not actually participate. [emphasis added]”⁶⁶ Further, no

⁶⁴ This is broadly analogous to the regulations and standards for research using identifiable health information. Both in New Zealand and overseas the emphasis is on pursuing research in a manner that will cause the least harm to the individual involved whether their consent is required or not. For the United Kingdom see Medical Research Council, Personal Information in Medical Research (January 2003) http://www.mrc.ac.uk/pdf- pimr.pdf. For New Zealand, see generally Davidson, Dawson and Moore, supra n. 35.

^{65 Manning [2004], supra n. 31, p. 400.}

⁶⁶ New Zealand Law Society, Submission on the Health (Screening Programmes) Amendment Bill (18 March 2003). http://www.nzlawsoc.org.nz/general/submissions/Health(Screening%20Progs)Amdt.ht m. (Emphasis added.)

benefit that would mitigate the intrusion of privacy accrues to the woman who has opted off the NCSP.

2. Selection of relevant information

Primary health care records frequently contain sensitive information about the individual. Although the Clinical Leader of the NCSP and the Director-General of Health have both stated that evaluators will only be interested in a woman’s smear history,⁶⁷ this limit is not found in the words of the statute. Evaluators are given full access to all information held by a hospital, laboratory or health practitioner on a relevant woman subject only to limits the Director-General may chose to impose at some stage, and the oversight of the woman’s doctor. Some writers have questioned the ability of evaluators to accurately identify all relevant information from patient files without “trawling through a woman’s life history”.⁶⁸ The dubious efficacy of the limits on evaluators reduces their value in being able to allay the privacy intrusion and so justify it. It also means the Amendment Act goes well beyond the standard expressed in the HIPC that in any case where information is released, “it must only be the minimum of information to secure the desired result.”⁶⁹

3. Obtaining consent

A crucial aspect of the case supporting the Amendment Act access provisions is the contention that it would be impossible to obtain consent from the necessary number of women to make an evaluation worthwhile. Some women may refuse access to their files and others will have died or otherwise not be traceable. This was a key concern about the national evaluation and cancer audit that the Gisborne Inquiry Committee sought. The intended time frame for consideration was the first decade of the NCSP’s operation so concerns about the ability to obtain the requisite level of consent are very reasonable. The same cannot be said for more contemporary evaluation and audit which

⁶⁷ See respectively Lewis, supra n. 10 and National Cervical Screening Programme, Sample Letter Important information about the National Cervical Screening Programme (21

February 2005). http://www.healthywomen.org.nz/NCSP/ncspPDFs/letter_QandA.pdf.

⁶⁸ McLean, ‘Dear Aunty Karen’, supra n. 30, p. 15.

^{69 Briscoe, supra n. 44.}

will become the norm under the Amendment Act. An Auckland University audit of cervical cancers diagnosed between 2000 and 2002 achieved very high response rates to consent for an interview and consent to access medical records: 85 percent of the women approached consented to at least one form of data collection, and 78 percent consented to all forms. The authors of the report note that their results indicate that “with appropriate resources and processes, it is possible to obtain high consent rates from both Mäori and non- Mäori women for interview and access to medical records.”⁷⁰

Professor McGoogan in her report on the NCSP also notes that obtaining individual women’s consent for participation in the audit has been relatively successful, though fairly slow.⁷¹

The results of the Auckland University audit also challenge the view that the participation of all relevant women is necessary for an evaluation or audit to be accurate and worthwhile. As noted earlier, a more complete data set produces more accurate results. The Amendment Act is clearly grounded in the belief that access to the histories of all women enrolled on the Register and those who develop cervical cancer is vital to an accurate and meaningful evaluation, a view shared by the Ministry of Health and Gisborne Inquiry Committee. The question of what is an appropriate level of accuracy for this kind of study is an epidemiological one. However, the measure of this is not. The level of accuracy must be necessary to meet the public health goal of evaluation of the NCSP and not merely be an additional bonus of an efficient data collection system.

If the Auckland University levels of consenting participation can be replicated in further evaluations and audits, and it can be shown that they produce meaningful and accurate evaluations, then a core justification for removing the consent requirement no longer exists. Further, and this point will be returned to below, it shows that there is another viable, practicable option that would obviously involve less harm to individual women.

⁷⁰ University of Auckland, The New Zealand Cervical Cancer Audit, supra n. 51,p. 3.

^{71 McGoogan, supra n. 3, [21].}

4. Informing women

The information provided to women about the Amendment Act changes could be seen as somewhat misleading as they are less than forthright about important details. The main contention is not that the information is inaccurate, but that it does not clearly or directly describe the scope of access that evaluators of the NCSP now have.⁷²

A letter sent to all women enrolled in the NCSP⁷³ is couched in the language of discretion and limitation: “evaluators may need to look at the personal health records of some women” to get information about what smears a woman has had, how often she has been tested, and the results of those smears. The background to the passage of the Amendment Act and the words of the statute clearly show however that evaluators will be looking at the health records of many women and that they have full access to any information contained therein that they deem relevant to their evaluation. Regulations that could limit this scope⁷⁴ have not been established and are unlikely to significantly curtail the powers of evaluators as this would defeat the purpose. In his submission on the Ministry of Health’s discussion document proposing the changes to the Health Act, the Privacy Commissioner queried the lack of clear and specific statements about what relevant information is, what evaluation involves and what exactly will be audited against what standards.⁷⁵ This lack of precision persists and tends to undermine the claim that women are adequately informed so as to justify or remove any invasion of their privacy.

5. The benefit to women

In general women will certainly benefit from a quality assured cervical screening programme. What is less clear is that there can be direct benefit from the use of their health records. If an evaluator considers a woman needs follow up treatment or becomes concerned about a health practitioner’s competence they may disclose the information to

⁷² See McLean, ‘Dear Aunty Karen’ supra n. 30, 15 and McGoogan, supra n. 3, [44]. Professor McGoogan notes that “institutional ‘speak’ and jargon is commonly used” by the NCSP and that communication with the public could be improved.

^{73 National Cervical Screening Programme, Sample Letter, supra n. 67.}

⁷⁴ Health Act 1956, section 112ZF.

^{75 Privacy Commissioner (NZ), supra n. 63.}

the relevant people⁷⁶ but there is no duty for them to act.⁷⁷ If the core premise of the argument in favour of the changes is that they will ultimately benefit women, it begs the question why evaluators are not obligated to act in these circumstances. This is especially surprising when viewed against the history of the NCSP.

6. Medical ethics

The medical practitioner’s duty to keep confidential the secrets confided in her is integral to the ethical basis of her practice, and to the patient-doctor relationship. Patients must trust in this confidentiality to ensure good health care. Medical ethics and the law do not always agree however, and in the case of NCSP evaluations there is a tension between the needs of epidemiologists and the ethics of clinicians. The New Zealand Medical Association remains concerned that if a woman who has opted off the NCSP and has developed cervical cancer refuses to allow evaluators access to her records, ethically her doctor ought to defy the Act.⁷⁸ In so doing, the doctor will be exposed to a fine of up to $10 000.⁷⁹ Medical professionals are left in an uncomfortable position between the basic duty of confidence they owe to their patients and the law requiring disclosure.

7. Alternatives

The overarching justification for the policy of the Amendment Act is that removing the requirement for additional consent from women is the only desirable and practicable way to meet the public health need to evaluate the NCSP. The existence of other viable options that involve a lesser incursion into individual women’s rights of privacy would severely diminish the weight of this argument. The obvious alternative is to reinstate the consent requirement and the University of Auckland Audit shows that this can work. Another possible solution is for the decision on whether consent is necessary or not to be left to ethics committees to decide on a case by case basis. This would involve the

⁷⁶ These are: the authority responsible for the registration of practitioners of the profession that the person concerned practises, the Accident Compensation Corporation or the Health and Disability Commissioner.

^{77 Health Act 1956, section 112Y.}

⁷⁸ Briscoe, supra n. 44, p. 7.

^{79 Health Act 1956, section 112ZP.}

newly established National Ethics Committee revising ethics committee guidelines to require consideration of the “moral obligations of audit” when making decisions.⁸⁰ While this cannot guarantee the participation of all eligible women in all audits or evaluations, ethics committee oversight is itself an important quality control mechanism in their design.⁸¹

Conclusion

The Health (National Cervical Screening Programme) Amendment Act

2004 highlighted for New Zealanders the tension between individual and community interests that is the central clash of any public health law. The individual right to confidentiality of personal health information is deeply entrenched, especially in the context of cervical screening. The controversy is that in this instance, the debate has been resolved in favour of the community interest in evaluation of the NCSP. Fundamentally, the underlying rationale of the Amendment Act is sound. Overriding consent requirements is the best way to guarantee a thorough audit of the NCSP and cervical cancer cases in order to make the screening programme better. Importantly however, the above analysis shows that there are key aspects of the policy that cannot be justified. Women who opt off the NCSP and develop cervical cancer are still subject to the Amendment Act despite an explicit decision not to participate, simply as a result of the disease they are diagnosed with. The emphasis on limits to the type of information that will be collected is rhetoric only with no specific grounding in the legislation. There are also significant flaws in the content of the justification arguments: obtaining consent from a sufficient number of women to make the evaluation meaningful is not necessarily unduly onerous; the information available to women is not candid about the use of their primary health care records or the specifics of opting off the register; evaluators are not obliged to act on evidence of practitioner incompetence or the need for further treatment; and, there are possibly viable alternatives that are less of an intrusion on women’s privacy.

⁸⁰ Paul, supra n. 53, p. 500.

^{81 Ministry of Health, Progress in Implementing the Recommendations, supra n. 48, [6.4].}

New Zealanders do not have a good understanding of public health screening programmes and what they necessarily entail.⁸² The Gisborne Cervical Screening Inquiry demonstrated that this has to change and evaluation and audit accepted as an integral part of screening programmes. This will not happen if the policy adopted appears to be unjustified and venture too far into the realm of personal autonomy. There are a number of significant legal and ethical concerns that underlie the public controversy about the Amendment Act and these ought to be addressed in order to meet the least harm standard and dispel women’s unease about this use of their personal health information. Even so, the Amendment Act provisions are grounded in concern for women’s health and they will ensure the quality of New Zealand’s National Cervical Screening Programme. Ultimately, the decision to allow the non-consensual use of women’s health records to facilitate this is the right decision.

⁸² McGoogan, supra n. 3, [4].