Unitas Hospital v Van Wyk and Another (231/2005) [2006] ZASCA 34 (27 March 2006)







THE SUPREME COURT OF APPEAL

OF SOUTH AFRICA


REPORTABLE

Case number : 231/2005


In the matter between :



UNITAS HOSPITAL APPELLANT


and


MARIA MAGDALENA VAN WYK FIRST RESPONDENT

DR G E NAUDÉ SECOND RESPONDENT




CORAM : HARMS, CAMERON, BRAND, CONRADIE et CLOETE JJA


HEARD : 23 FEBRUARY 2006


DELIVERED : 27 MARCH 2006


Summary: Access to information – s 50 of Promotion of Access to Information Act 2 of 2000 – meaning of ‘required’ for exercise of protection of right – when available to compel pre-action production. (Order in para 27.)


Neutral citation: This judgment may be referred to as Unitas Hospital v Van Wyk [2006] SCA 32 (RSA)


JUDGMENT


BRAND JA/


BRAND JA:

[1] This appeal has its origin in an application by the first respondent, Mrs van Wyk, against the appellant hospital (‘Unitas’) and the second respondent, Dr Naudé, under the provisions of the Promotion of Access to Information Act 2 of 2000 (PAIA) in the Pretoria High Court. The background to the application appears from what follows. Unitas is a private hospital in Centurion. On 3 July 2002, Mrs van Wyk’s late husband died while he was a patient in Unitas. Mrs van Wyk contended that the death of her husband was brought about by the negligence of Unitas’s nursing staff. She also believed, on the basis of expert evidence received, that she had an action against Unitas for the damages that she suffered through her late husband’s demise. Her application for access to information was with a view to the institution of that action.


[2] Dr Naudé is a specialist physician with rooms at Unitas. He was one of the doctors who treated the deceased during the term of his admission. Apart from his private practice, Dr Naudé served as director of the multi-intensive care unit (‘the ICU’) at Unitas. He was also the chairman of the body representing the medical specialists practising at Unitas in their dealings with the hospital management. Acting in these last mentioned capacities, he prepared a report on general nursing conditions in the ICU and the high care unit at Unitas. In the court a quo this report, which had been finalised on 28 June 2002, was referred to as ‘the Naudé report’. I propose to use the same terminology.


[3] In terms of s 50 of PAIA, private bodies, such as Unitas, and individuals in professional practice, such as Dr Naudé, are obliged to provide a ‘requester’ with recorded information in their possession if the requirements of s 50 are met. Initially Mrs van Wyk’s application under the provisions of this section was for access to a number of documents. However, as matters developed on the papers, the issues between the parties eventually turned solely on whether she was entitled to a copy of the Naudé report. While Unitas opposed that application, Dr Naudé filed an answering affidavit in which he indicated that he neither opposed nor consented to the order sought against him, but that he abided the decision of the court. The court a quo (Mojapelo J) granted the application against Unitas with costs. At the same time it gave what was described as an ‘order by default’ against Dr Naudé. In the event, both Unitas and Dr Naudé were ordered to provide Mrs van Wyk with copies of the Naudé report. This appeal, which only relates to the order granted against Unitas, is with the leave of this court.


[4] According to s 50 of PAIA a ‘requester’ is entitled to the records – which, by definition, include any recorded information – in the possession of a private body if:

(a) the information ‘is required for the exercise or protection of any rights’;

(b) the ‘requester’ complies with the procedural requirements of PAIA relating to a request for access to that information;

(c) access to the information requested cannot be refused on any of the grounds provided for in ss 62 to 70.


[5] One of the objections initially raised by Unitas was that divulgence of the Naudé report would be contrary to s 63(1) of PAIA in that it ‘would involve the unreasonable disclosure of personal information about’ members of Unitas’s nursing staff who were criticised in the report by name. The cause for this objection was, however, removed when Mrs van Wyk conceded that the copy of the report provided to her could be edited by blanking out the name of any third party whose privacy would otherwise be infringed by disclosure. This concession was, incidentally, given effect to in the order eventually granted by the court a quo. In the result, Unitas no longer maintained that access to the Naudé report would be contrary to any of the provisions of ss 62 to 70. Nor was it contended that Mrs van Wyk did not comply with any of the procedural requirements of the Act. Unitas’s remaining ground of opposition was therefore that Mrs van Wyk had failed to establish that she required the Naudé report for the exercise or protection of any right.


[6] A substantial part of the argument in this matter had been devoted to the meaning of ‘required’ within the context of s 50(1)(a). I shall come to that. It seems plain, however, that any attempt to determine its meaning in the abstract would be a futile exercise. Generally speaking, the question whether a particular record is ‘required’ for the exercise or protection of a particular right is inextricably bound up with the facts of that matter. Accordingly, I revert to the facts.


[7] Mrs van Wyk’s late husband suffered from Chrone’s disease. He was admitted to Unitas on 6 May 2002 where he was surgically operated upon on 13 May 2002. After the operation he went to the ICU. Three days later he was transferred to the high care unit but, because he subsequently developed an infection, he went back to the ICU. From then on Mr van Wyk’s condition remained serious and he essentially alternated between the ICU and the high care unit. On 1 June 2002, and while under care of nursing sisters, he suffered an attack of vomiting and aspirated. After that, he remained in the ICU where he died on 3 July 2002.


[8] During his period of admission to Unitas, Mr van Wyk was treated by various medical specialists, including Dr Naudé. Mrs van Wyk visited her husband every day. She was overtly critical of the nursing staff and the conditions in the hospital – so much so that at some stage she arranged for a representative of her husband’s medical aid fund to visit and assess the treatment that he was receiving. Her objections were pursued in a detailed letter of complaint that she submitted to Unitas after the death of her husband, as well as in a subsequent letter of demand written by her attorneys on her behalf. In these two letters and in her founding affidavit Mrs van Wyk repeated her general allegations that conditions in the hospital were unhygienic and that the nursing staff were incompetent, improperly trained, unsympathetic and uncooperative. It is not clear, however, how these general allegations are linked to her late husband’s death. Her more pertinent complaint related to the incident on 1 June 2002 when the deceased suffered an attack of vomiting and aspirated. According to Mrs van Wyk this occurrence, which allegedly led to his death, happened because the deceased was given solid food to eat against the instructions of his doctor, and was not properly supervised.


[9] In the letter of complaint, Mrs van Wyk informed Unitas that she intended instituting action against it and that she was assisted in the formulation of her claim by seven medical specialists. Six of these experts, including Dr Naudé, were involved in the treatment of her late husband during the period of his admission to Unitas, while the seventh is a medico-legal expert who is both a qualified orthopaedic surgeon and an admitted advocate.


[10] With specific reference to the Naudé report, Mrs van Wyk alleged in her founding affidavit that the report was the product of an investigation into nursing conditions at Unitas, which was undertaken specifically as a result of her husband’s death. These observations obviously reflected her belief at the time. From the answering affidavits it appeared, however, that that belief was mistaken. According to both Unitas and Dr Naudé, the report was completed prior to the death of the deceased on 28 June 2002 and it was not motivated, as Mrs van Wyk thought, by the death of her late husband, but because of what Dr Naudé described as a ‘general disquiet with the quality and efficiency of the nursing care in the ICU and high care unit of the hospital’. The report was the product of a survey in these two units over a period of six days from 18 to 24 June 2002. The survey was undertaken at Dr Naudé’s initiative, but with the consent of the hospital management. It was done by three ICU qualified sisters over the six day period and was intended as a general review of existing practices of nursing care in the two units concerned. It included an assessment of matters such as inter-personal problems and personality differences at the level of nursing management; and the ability, training and remuneration of the nursing staff. The report resulting from the survey offered an assessment of the quality of patient care, staff morale and management issues in general. It contained no specific reference to the treatment of the deceased. Indeed, so both Unitas and Dr Naudé stated, the name of the deceased is not even mentioned in the report. The fact that the survey was conducted during part of the period of the deceased’s admission to the two units concerned, they said, was purely coincidental.


[11] In her founding papers, Mrs van Wyk did not specifically state that she required the Naudé report for the exercise of any right. What she said was that, without access to the report, her right to claim damages from Unitas would be affected (‘aangetas word’). Maybe because of her belief that the report was directly linked to the death of her husband, she did not elaborate on what benefit she thought she could derive from the contents of the report. In their answering affidavits, both Unitas and Dr Naudé squarely raised the defence that Mrs van Wyk did not ‘require’ the Naudé report for the exercise or protection of any rights, as contemplated by s 50 of PAIA. This is so, they said, essentially because she already had access to whatever information her experts would require to advise her on the formulation and the assessment of her claim.


[12] In motivating this defence, Unitas and Dr Naudé pointed out, first, that because the report was of a general nature, Mrs van Wyk was mistaken in her belief that it had any specific bearing on the treatment received by her husband; secondly, that it had been recorded by Mrs van Wyk in her letter of complaint that she was assisted by seven medical specialists, including Dr Naudé himself, who was the author of the report; thirdly, that, apart from her own recorded observations regarding the treatment received by her late husband, Mrs van Wyk had already been provided with a complete set of his hospital records, including the notes made by the nurses caring for him throughout his time in hospital, recording all their observations as well as the treatment that he received. Moreover, they said, the medical experts advising Mrs van Wyk had access to all the clinical notes and medical reports of the doctors responsible for his treatment. In these circumstances, Unitas stated, Mrs van Wyk had

all the information which her experts would require in order to advise her’

while Dr Naudé said:

[T]he applicant should be more than capable of procuring the relevant facts pertaining to her husband’s hospitalisation from the contemporaneous records which specifically relate to him’,

and went even further by saying:

[B]eyond attempting to embarrass [Unitas], there is nothing to be gained concerning the hospitalisation and treatment of [her] spouse from the report.’


[13] In reply, Mrs van Wyk did not deny any of these specific allegations by the deponents to the answering affidavits. Apart from the general statement that she required the report for the exercise and protection of her right to claim damages, she essentially said three things. First, that because the report related to the general nursing conditions in the two units of the hospital where her husband had been treated and over part of the period during which he received such treatment, it must, as a matter of inference, be relevant to the question whether the nursing staff had been negligent or not. Secondly, that if it is so that the disclosure of the report would tend to embarrass Unitas, it could only be because the general nursing care in the two units concerned had not been up to standard, a conclusion which would in turn support her case. Thirdly, that the report would assist her in establishing whether she had reasonable prospects of success on the merits of the case.


[14] On these facts the court a quo found that Mrs van Wyk had succeeded in substantiating her requirement of the Naudé report for the exercise and protection of her right to claim damages. The nub of its reasoning in support of this finding appears to be incorporated in the following quotations from the court’s judgment:

[I]t seems to me that having access to the Naudé report will assist the applicant in either proceeding with or abandoning the claim against the first respondent. It is in the interest of bona fide litigation that the parties should take this critical decision on the basis of essential information where such is available. In the words of Cameron J [in Van Niekerk v Pretoria City Council 1997 (3) SA 839 (T) 848G]:

Either way, disclosure will promote an early settlement of the dispute and bring the envisaged litigation, by settlement or abandonment, to a short, sharp end. In this sense, the applicant can in my view be said reasonably to require the report.”

Such is indeed the view I hold in the present case.’

And:

While the applicant may arguably be in possession of some information on the basis of which she may issue summons against the first respondent, there can be no doubt that the less the information she has the higher the risk she runs of either formulating her claim incorrectly or even of proceeding with the litigation when, with a fuller picture, she might decide not to. . . . Another advantage of assessing information early is that, if indeed based on the available information the applicant does not have a case against the first respondent, the applicant will be placed in the position to know this early and therefore avoid unnecessary litigation. This, as I understand the position, is precisely part of the driving notion behind the ideals of promoting access to information. Access to information must limit prejudice and encourage or facilitate early or timeous resolution of disputes. It is a fitting philosophical approach to dispute resolution in an open and democratic society.’


[15] In the course of its judgment the court a quo referred to a number of reported decisions in which attempts were made to determine the meaning of ‘require’ in the context of the phrase ‘require for the exercise or protection of any right’. Though most of these decisions were not concerned with s 50 of PAIA, specifically, but with the identical phrase used in s 23 of the 1993 Constitution and in s 32 (read with Schedule 6) of the 1996 Constitution, they obviously have a bearing on the understanding of the same expression in s 50. It therefore comes as no surprise that essentially the same decisions were referred to in the judgment of this court concerning the interpretation of s 50, which was delivered after the court a quo had given its judgment, in Clutchco (Pty) Ltd v Davis 2005 (3) SA 486 (SCA) paras 11 and 12.


[16] Apparent from all the decisions referred to is the reluctance of the courts involved to make any positive statements as to what the expression ‘require’ means. The inclination is rather to define the expression in terms of what it does not mean. So, for example, it is said that it does not mean the subjective attitude of ‘want’ or ‘desire’ on the part of the requester; that, at the one end of the scale, ‘useful’ or ‘relevant’ for the exercise or protection of a right is not enough, but that, at the other end of the scale, the requester does not have to establish that the information is ‘essential’ or ‘necessary’ for the stated purpose (see eg Khala v Minister of Safety and Security 1994 (4) SA 218 (W) at 224G-225E; Shabalala v Attorney-General, Transvaal 1995 (1) SA 608 (T) at 624C; Nortje v Attorney-General, Cape 1995 (2) SA 460 (C) 474G). Closest to a positive formulation is the one articulated as follows by Streicher JA in Cape Metropolitan Council v Metro Inspection Services (Western Cape) CC 2001 (3) SA 1013 (SCA) para 28:

Information can only be required for the exercise or protection of a right if it will be of assistance in the exercise of protection of the right. It follows that, in order to make out a case for access to information . . . an applicant has to state what the right is that he wishes to exercise or protect, what the information is which is required and how that information would assist him in exercising or protecting that right.’


[17] The threshold requirement of ‘assistance’ has thus been established. If the requester cannot show that the information will be of assistance for the stated purpose, access to that information will be denied. Self-evidently, however, mere compliance with the threshold requirement of ‘assistance’ will not be enough. The acceptance of any notion to the contrary will, after all, be in conflict with the postulate that mere usefulness to the requester will not suffice. In Clutchco this court was reluctant to go any further than to confirm this threshold requirement. That appears from the following statement by Comrie AJA immediately after he had referred to the above quoted dictum from Cape Metropolitan Council (in para 13):

I think that ‘reasonably required’ in the circumstances is about as precise a formulation as can be achieved, provided that it is understood to connote a substantial advantage or an element of need. It appears to me, with respect, that this interpretation correctly reflects the intention of the legislature in s 50(1)(a).’


[18] I respectfully share the reluctance of Comrie AJA to venture a formulation of a positive, generally applicable definition of what ‘require’ means. The reason is obvious. Potential applications of s 50 are countless. Any redefinition of the term ‘require’ with the purpose of restricting its flexible meaning will do more harm than good. To repeat the sentiment that I expressed earlier: the question whether the information sought in a particular case can be said to be ‘required’ for the purpose of protecting or exercising the right concerned, can only be answered with reference to the facts of that case having regard to the broad parameters laid down in the judgment of our courts, albeit, for the most part, in a negative form.


[19] With regard to the facts of this case, it can be accepted with confidence that Mrs van Wyk did not require the Naudé report to formulate her claim for the purposes of instituting an action. What must also be accepted is that, once she has instituted her action and provided that the Naudé report turns out to be relevant to the issues on the pleadings, Unitas will be obliged to make it available under the provisions for discovery in terms of Uniform rule 35. From the provisions of s 7 of PAIA it is plain, in my view, that PAIA is not intended to have any impact on the discovery procedure in civil cases. Once court proceedings between the parties have commenced, the rules of discovery take over. In that event, access to documents in possession of the litigating parties is governed by these rules. The provisions of PAIA no longer apply as between the parties (see eg Ian Currie & Jonathan Klaaren The Protection of Access to Information Act Commentary, para 4-15 at 52-54. Cf also Inkatha Freedom Party v Truth and Reconciliation Commission 2000 (3) SA 119 (C)).


[20] The real issue is therefore, whether in the circumstances of this case, s 50 afforded Mrs van Wyk a right to what would amount to pre-action discovery. The court a quo concluded that it did. According to the passages from the court’s judgment quoted in para [14] above, its motivation for this conclusion was that the report could possibly assist her in establishing the merits of her case, which would in turn enable her to decide whether she should embark on the risky venture of litigation at all. What is more, it appears from these passages that the court a quo was of the view that underlying considerations such as avoidance of speculative litigation and early determination of disputes would, as a matter of course, entitle a requester to information under s 50 if such information could possibly be of assistance in establishing the merits of his or her case. That is so, the court held, because these underlying considerations would accord with the ‘philosophical approach to dispute resolution in an open and democratic society’.


[21] I find myself in respectful disagreement with these sentiments. I do not believe that open and democratic societies would encourage what is commonly referred to as ‘fishing expeditions’ which could well arise if s 50 is used to facilitate pre-action discovery as a general practice (see Inkatha Freedom Party (supra 137C). Nor do I believe that such a society would require a potential defendant, as a general rule, to disclose his or her whole case before any action is launched. The deference shown by s 7 to the rules of discovery is, in my view, not without reason. These rules have served us well for many years. They have their own built-in measures of control to promote fairness and to avoid abuse. Documents are only discoverable if they are relevant to the litigation while relevance is determined by the issues on the pleadings. The deference shown to discovery rules is a clear indication, I think, that the legislature had no intention to allow prospective litigants to avoid these measures of control by compelling pre-action discovery under s 50 as a matter of course.


[22] I hasten to add that I am not suggesting that reliance on s 50 is automatically precluded merely because the information sought would eventually become accessible under the rules of discovery, after proceedings have been launched. What I do say is that pre-action discovery under s 50 must remain the exception rather than the rule; that it must only be available to a requester who has shown the ‘element of need’ or ‘substantial advantage’ of access to the requested information, referred to in Clutchco, at the pre-action stage. An example of such a case is, in my view, to be found in Van Niekerk v Pretoria City Council (supra), upon which considerable reliance was placed by the court a quo (see the quotation in para [14] above). The point is, however, that the facts of that case were materially different. Van Niekerk had a report by experts who did not identify who was responsible for the damage to his equipment (848C). The City Council, on the other hand, relied on a report which apparently exonerated it from responsibility (848F-G). Quite understandably, in the circumstances, Van Niekerk’s allegation was that without the report relied upon by the City Council, he was unable to establish whether it could be held liable (848H-I). Though I think it is legitimate to use s 50 to identify the right defendant, I do not agree with the court a quo’s thesis that one is entitled, as a matter of course, to all information which will assist in evaluating your prospects of success against the only potential defendant. On that approach, the more you know the better you will be able to evaluate your chances against your opponent. The corollary of this thesis therefore seems to be that the requester will in effect always be entitled to full pre-action discovery. The dicta by Cameron J in Van Niekerk referred to by the court a quo (see para [14] above) cannot legitimately be relied upon in support of its thesis.


[23] Mrs van Wyk is not in the same category as Van Niekerk. On her own showing she had a number of alternate sources of information available to her as to what happened to her late husband while he was in Unitas. In the circumstances, she should, in my view, have explained from the outset what more knowledge she hoped to gather and what benefit she hoped to attain by gaining access to the Naudé report. No such allegation was, however, made in her founding affidavit. But it goes further. In their answering affidavits, both Unitas and Dr Naudé made the positive statements that she already had access to all the information her experts needed to advise her and that the Naudé report would add nothing to her case. These statements were not denied in reply. Even when Dr Naudé said categorically that the only advantage Mrs van Wyk could gain from his report would be to embarrass Unitas, this was not denied. Mrs van Wyk’s reply was that such embarrassment could only result from sub-standard nursing which would in turn support her case. In my view, this is a non sequitur. From the nature of the survey that preceded the report, as it is explained in the answering papers, I can think of a number of ways in which the report can embarrass the hospital without having any bearing on Mrs van Wyk’s case at all. Mrs van Wyk’s further contention in reply, that, because the report resulted from a survey in the units where her late husband had been treated, it must be inferred to be relevant, is clearly not enough.


[24] It is true, as was argued on behalf of Mrs van Wyk, that she was at a disadvantage to show the relevance of the report without any knowledge of its contents. That, of course, would normally be the position of any requester who seeks information to which he or she had no prior access. What makes this case somewhat different is that Dr Naudé, who was the author of the report, is also one of Mrs van Wyk’s advisors. In fact, it is common cause that, but for Unitas’s objection, he would have given her the report at the outset. The question remains, however, why Dr Naudé could not be asked to provide her with all the information available to him. To this question there is no answer on the papers. The mere fact that some of this information might also be reflected in the report would not, in my view, make any difference.


[25] In the circumstances I conclude that Mrs van Wyk had failed to substantiate her claim that the Naudé report would be of assistance to her in her case against Unitas. It follows that she did not even meet the threshold test formulated in Cape Metropolitan Council (see para [16] supra). The question whether in addition she had shown the ‘element of need’ or the ‘substantial advantage’ suggested in Clutchco (see para [17] supra), therefore does not even arise. Consequently I hold the view that her application against Unitas should have been dismissed.


[26] Lastly, an alternative argument was raised on behalf of Mrs van Wyk which was really in the nature of a point in limine based on s 21A of the Supreme Court Act 59 of 1959. In essence it went as follows: Since Dr Naudé had already been ordered to provide Mrs van Wyk with the same report and since there is no appeal against that order, the appeal by Unitas (in the words of s 21 A) could have no practical effect or result and should therefore not be entertained. It appears, however, that thus far no attempt has been made on behalf of Mrs van Wyk to compel compliance with the order against Dr Naudé. It therefore seems to have been accepted by everybody concerned, at least by implication, that that order would follow the same fate as the one against Unitas. In fact, the order against Dr Naudé was granted by the court a quo on the basis that the order against Unitas was in any event to ensue. In the circumstances, the refusal to hear Unitas’s appeal merely because a similar order was technically enforceable against Dr Naudé, would, in my view, be unjustified.


[27] The appeal is upheld with costs, including the costs of two counsel. For the order by the court a quo there is substituted the following:

The application is dismissed with costs, including the costs of two counsel.’




………………..

F D J BRAND

JUDGE OF APPEAL


CAMERON JA:


[28] I have had the benefit of reading the judgment of my colleague Brand JA, but regret that I cannot agree with it. We differ in our approach not only to the details of the case, but to the principles that underlie it. The question is whether Mrs van Wyk is entitled to obtain from the hospital a physician’s report that was compiled in the month before her husband died in the hospital in July 2002, and which dealt in general terms with the nursing situation in its multi-intensive care unit and high care unit at a time when he was a patient in the multi-ICU.


[29] Mrs van Wyk acknowledges that she does not need the report to get her damages claim against the hospital off the ground – as a matter of bare pleading, she can get by without it.1 She may be entitled to receive it later, when pre-trial exchange of documents takes place under the rules of discovery. But she claims it now, and she is entitled to it if she requires it for the exercise or protection of her rights. In my view, for substantially the reasons set out in the judgment of Mojapelo J in the court below, she does. I respectfully differ from my colleague’s conclusion to the contrary.


[30] The Promotion of Access to Information Act 2 of 2000 (PAIA) provides that access must be given to records held by private bodies if the document requested ‘is required for the exercise or protection of any rights’ (s 50(1)(a)). The word ‘required’ originated in the interim Constitution, under which the right of access applied only against organs of state.2 The Bill of Rights extended this right horizontally to ‘any information that is held by another person and that is required for the exercise or protection of any rights’.3 The meaning of ‘required’ is neither clear nor precise, and judges have grappled for more than a decade to give it a practical content. As Brand JA explains, this court eventually settled on a pragmatic formulation in Clutchco (Pty) Ltd v Davis:4 a private body must yield a record it holds if the applicant can show ‘substantial advantage or an element of need’. But it is important to bear in mind that in adopting this standard Clutchco expressly rejected the notion that ‘required’ meant ‘necessity’ – still less ‘dire necessity’. Like the statute, the standard is accommodating, flexible and in its application necessarily fact-bound.


[31] In applying the standard to any set of facts, we must be faithful to the objects of PAIA and the constitutional setting from which it springs. In thus considering whether Mrs van Wyk has demonstrated an element of need in relation to the report, or has shown that having it will give her a substantial advantage, I differ from Brand JA in three respects. In summary:

(a) The nature of Mrs van Wyk’s need and the substantial advantage the report will afford her: It is in my view too narrow to approach Mrs van Wyk’s PAIA entitlement by focusing on whether she is capable of formulating a damages claim against the hospital. Her litigation against the hospital will involve issues not only of individual conduct and responsibility. It will scrutinise how the hospital’s systems functioned, and whether institutional procedures and protocols caused or permitted actions by individual members of staff. She is entitled to the report because knowing what it says will afford her a significant advantage in relation to those questions.

(b) The approach to private bodies under PAIA: Though the hospital is not a public body under PAIA, the statute suggests that our approach to private bodies should not be undifferentiated. Some bodies are more private than others. The hospital is a lot less private than the small family business in Clutchco. The declared objects of PAIA5 suggest that where appropriate courts should encourage transparency, accountability and effective governance in private institutions: large private institutions that serve the public and, like the hospital, perform vital public functions fall plainly within the scope of this statutory objective. The judgment of Brand JA in my respectful view goes in the opposite direction.

(c) Pre-litigation discovery: In my respectful view the approach Brand JA propounds is too cloistered. Pre-discovery disclosure is important and helpful in assisting a litigant – and thereby also the opponent – to determine whether litigation should commence at all, or whether it should proceed. PAIA recognises the importance of post-commencement access procedures; but its novel dimension lies in the fact that it creates pre-commencement access. We should not stifle this. Litigation involves massive costs, time, personnel, effort and risks. Where access to a document can assist in avoiding the initiation of litigation, or opposition to it, the objects of the statute suggest that access should be granted.


The nature of Mrs van Wyk’s need and the substantial advantage the report will afford her

[32] Mrs van Wyk thought that the report of Dr Naudé dealt amongst other things with her husband’s care and treatment while he lay critically ill in the hospital. She was wrong. In his answering affidavit, Dr Naudé explains that the survey was conducted ‘purely coincidentally’ at the time when Mrs van Wyk’s husband was in the hospital. It was not prompted by nor did it relate to ‘the nursing care which [Mrs van Wyk’s] spouse specifically received’ at the hospital, and it does not refer to him. Instead, by virtue of ‘general disquiet with the quality and efficiency of the nursing care’ in the hospital’s multi-intensive care unit and high care unit, Dr Naudé says he undertook ‘an informal survey of the nursing care’ in these units in June 2002. It was performed over a period of about six days and was conducted by three intensive care unit trained sisters under his supervision. It included assessments of ‘a wide variety of aspects of nursing care in the units’, and embraced, Dr Naudé tells us –

(i) a general review of the implementation of nursing protocols;

(ii) a general review of the existing practices of nursing care;

(ii) an assessment of the adequacy of staffing;

(iv) an assessment of inter-personal problems and personality differences between nursing management and the unit managers;

(v) an assessment of the appointment of nursing staff;

(vi) an assessment of the abilities of nursing staff allocated to the wards in question;

(vii) an assessment of the adequacy of continued nursing education of nursing personnel;

(viii) an assessment of the adequacy of nursing staff remuneration.


[33] The survey was intended, Dr Naudé says, ‘to serve as a tool for achieving and maintaining high standards of nursing care in the units in question’. Rather than being scientific and comprehensive, it was ‘an information survey and report, aimed at achieving and maintaining high standards of nursing care’.


[34] Can Mrs van Wyk against this background be said to require the report for the exercise and protection of her rights? That depends on how one approaches her claim. S he avers that the hospital negligently caused her husband’s death. To institute action to vindicate that entitlement, all the rules of pleading require her to do is to set out particulars of the individual act or acts of negligence she claims gave rise to her loss. But Mrs van Wyk’s founding affidavit (which was drafted when she was still considering whether to institute action) explains her claim in far broader terms:6

During the deceased’s long hospitalisation it was clear that there were various problems at [the hospital]. There were for example extremely unhygienic conditions. It was also clear that some of the personnel were not trained to manage the ward to which they were assigned. There were also insufficient personnel present and there were often only unqualified and under-qualified personnel on duty;

As a result of food that the deceased was given to eat by [the hospital’s] personnel, against the express instructions of the treating doctors, and as a result of insufficient supervision in the side-room of intensive care, where the deceased was being nursed at that stage, the deceased aspirated and later in consequence died.’ (My translation.)


[35] If one views this claim as encompassing only the individual acts of negligence that the rules of pleading require her to set out in order to launch a viable action, it is possible to conclude that Mrs van Wyk does not ‘require’ the Naudé report for the exercise and protection of her rights, and counsel for the hospital urged us to find that that the test was whether she was in a position to commence her action against the hospital without it. But this is far too narrow. And in my view it is in any event quite wrong to view Mrs van Wyk’s claim as encompassing only isolated or unconnected acts of negligence. She does not claim that an individual actor, negligent alone, caused her husband to die. What she claims is that an institution – the hospital – failed in its functioning in ways that gave rise to the individual acts of negligence that caused her husband to die. She claims that a systemic, not individual, failure caused her loss.


[36] To view her claim as traversing only the individual acts of negligence is to divorce those acts from the institutional setting that gave rise to them. It is to divorce them from the systems that she alleges not only permitted them to occur, but whose insufficiency she claims lay at the heart of the individual acts of sub-standard nursing that caused his death.


[37] Her founding affidavit’s statement of case traverses much of the very ground that Dr Naudé says his report covers (though he explains that hygiene was not part of it). So seen, it is clear that the report would have a material and direct bearing on the whole basis of her claim. My colleague Brand JA concludes (para 23) that the report might embarrass the hospital without having any bearing on Mrs van Wyk’s case at all. I respectfully differ. It seems to me that the report will illuminate the core of Mrs van Wyk’s claim, which is that an institution to which she and her husband entrusted his health failed in its essential functioning, so causing his death and her loss. In this it will assist her materially in both formulating and evaluating her claim.


[38] While Dr Naudé’s report does not mention her husband’s specific case or details, it is not hard to surmise, from the doctor’s exposition of its contents, that its every part will bear on the background to and causes of the institutional and systemic failures she alleges were fatal – and which Dr Naudé was monitoring in the very days before his death. Having the Naudé report will therefore confer a substantial advantage on her. It will set out for her the very protocols and practices that were in place while her husband lay mortally ill; the problems – both in individual personality and in remuneration and training – that she claims contributed to the fatality; and the appointment procedures and standards whose existence permitted the fatal failures to occur. It will cast light on the very organisational issues that led to the malfunction that she claims caused her loss. By doing so it will enable her not only to particularise her claim, but to assess its ambit and its viability.


[39] In my respectful view, Mrs van Wyk established the requisite statutory element of need and is entitled to the report.


The approach to private bodies under PAIA: how private is the hospital?

[40] Following the distinction Bill of Rights draws between information held by the state and that held by other persons, PAIA distinguishes between public and private bodies, each of which it defines. In the case of the former, there is a general right of access. In the case of the latter, access must be required for the exercise or protection of rights. But, as Brand JA explains, ‘required’ is a flexible term, and its application must be fact-bound. And in applying it to any particular case, we must in my view consider the extent to which it is appropriate in the case of any private body to further the express statutory object of promoting ‘transparency, accountability and effective governance’ in private bodies. This statutory purpose suggests that it is appropriate to differentiate between different kinds of private bodies. Some will be very private, like the small family enterprise in Clutchco. Effective governance and accountability, while important, will be of less public significance. Other entities, like the listed public companies that dominate the country’s economic production and distribution, though not ‘public bodies’ under PAIA, should be treated as more amenable to the statutory purpose of promoting transparency, accountability and effective governance.


[41] We have little on record about the hospital’s corporate setting, control or ownership. What we do know from the answering affidavits is that is a health facility offering nursing and other services to the general public, and that it does so as part of the Netcare group (which, it is well-known, has clinics country-wide and is one of a very small number of publicly listed corporations that dominates the field of private healthcare services). We also know that the range of doctors practising from the hospital’s premises is large enough to warrant the establishment and functioning of a representative body called the ‘Hospital Board’, which Dr Naudé chairs. We know finally that the hospital has a substantial and diverse nursing complement, for in response to Mrs van Wyk’s access request it lists no fewer than 36 nurses and managers whose confidentiality could have been affected by the Naudé report’s disclosure.


[42] The hospital is therefore a rather public private body, and in dealing with Mrs van Wyk’s request for access to the report we are obliged to give effect to the objects of the statute. These include, generally, the promotion of ‘transparency, accountability and effective governance of all public and private bodies’. This provision seems to me very much to have in its focus private bodies like the hospital, which offers essential services to the public, yet is subject to no direct public or political accountability in its conduct or governance. The statute enjoins us to promote transparency and accountability and effective governance in the hospital, and in interpreting whether Mrs van Wyk requires the report for the exercise and protection of her rights we should in my view be astute not to help it shroud its institutional weaknesses and failures from pre-trial scrutiny.

Pre-litigation discovery: the opportunity for beneficial change

[43] PAIA does not apply to records requested for criminal or civil proceedings after the commencement of proceedings where ‘the production of or access to that record for the purpose [of the proceedings] is provided for in any other law’ (s 7). The effect of this provision is that where an applicant/plaintiff is entitled to obtain a document by discovery, the statute’s provisions do not apply. My colleague Brand JA does not suggest that the provision precludes access simply because the information sought will eventually become accessible under the rules of discovery (para 22). But he cautions against letting pre-trial discovery become the rule rather than the exception, and he considers that Mojapelo J erred in permitting Mrs van Wyk access to the Naudé report on the basis that it would assist her in evaluating her prospects of success against the hospital.


[44] I respectfully differ from my colleague’s approach. Like him, I do not consider that PAIA offers untrammelled pre-action disclosure. But I do not think that this is what Mojapelo J had in mind, or what his order in favour of Mrs van Wyk portended. In particular, I disagree distinctly with my colleague’s apparent suggestion (para 21) that Mojapelo J’s approach, or recognising Mrs van Wyk’s claim, would be to license fishing expeditions. In my view, Mrs van Wyk has established a clear and substantial connection between her claim against the hospital and the contents of the report she seeks to see.


[45] What underlies our difference is a broader question: the approach to the impact of the statute on the old order of pleadings and action. In my view, as Mojapelo J rightly suggested, the statute affords an opportunity to broaden the approach to pre-action access. It does so on a basis that is flexible and accommodating without threatening the boundless exposure against which my colleague warns. The key lies in a case-by-case application of whether a litigant ‘requires’ a record.


[46] PAIA’s scheme is to afford access before litigation, but then to withdraw its procedures after the commencement of action. This differs from other jurisdictions. In England, for instance, the Freedom of Information Act of 2000 applies only to ‘public authorities’, and there is no pre-litigation access under it to the records of private bodies. But the rules of court promulgated under the Supreme Court Act of 1981 permit pre-action disclosure where there is a real prospect in principle that this will be fair to the parties if litigation is commenced, or of assisting them to avoid litigation, or to avoid costs in any event.7


[47] In my view, it is a legitimate and beneficial approach to the statute to apply it so that where a record will assist a party in evaluating a potential claim, this will count as advantage or need for statutory purposes. This was what lay behind the reasoning, as opposed to the particular facts, in van Niekerk v Pretoria City Council:8 the objective of promoting early settlement of disputes by assisting potential litigants to evaluate the viability of their claims in the light of the documentation sought. It was this spirit that Mojapelo J endorsed in approaching the facts of this case, so recognising Mrs van Wyk’s entitlement.


[48] Institution of proceedings is an immense step. It involves a massive commitment in costs, time, personnel and effort. And it is fraught with risks. Where access to a document can assist in avoiding the initiation of litigation, or curtailing opposition to it, the objects of the statute suggest that access should be granted. On this consideration, too, Mrs van Wyk is entitled the report.


[49] I would therefore dismiss the appeal with costs.




E CAMERON

JUDGE OF APPEAL





CLOETE JA:


[50] I have had the advantage of reading the judgments of my brothers Brand and Cameron. I agree with the reasoning and conclusion in the former and wish to state the reasons why I cannot, with respect, subscribe to the approach followed in the latter with regard either to the interpretation of PAIA or to the assessment of Mrs van Wyk’s case.


[51] Under PAIA a requester either has a right to know what is in a record or must demonstrate that the record is required. The distinction depends upon the nature of the body to which the request is addressed. If it is a public body, s 11 applies and access to a record must be given if the procedural requirements have been fulfilled and if access is not refused on a recognised ground of refusal. If it is a private body, s 50(1) imposes a further requirement, namely, that the record is ‘required for the exercise or protection of any rights’. The answer to the question whether the record is so required must be sought by having regard to the position of the requester; the answer does not depend in any way on the classification of the body to which the request is addressed. That classification determines whether the question has to be asked, not how it should be answered. Either a body is a public body or it is a private body as defined in the statute (or it is neither, in which case the statute finds no application) and there is, with respect, no warrant for describing a body as a ‘rather public private body’ nor is its size or the type of trade, business or profession in which it is engaged, relevant. The stated objectives of the statute cannot be utilized selectively to erode the fundamental distinction which the provisions of the statute in terms establish. I say selectively, because one of the objects of PAIA (set out in s 9(b)(i)) is to give effect to the constitutional right of access subject to justifiable limitations, including limitations aimed at the reasonable protection of privacy and confidentiality. Furthermore s 32(1) of the Constitution itself draws a clear distinction between any information held by the State and any information that is held by another person.


[52] The information that Mrs van Wyk had, is set out in para 12 of the judgment of my brother Brand. Some of that information was known to her personally; the rest had been provided to her or was available to her. In particular, the author of the report, Dr Naude himself, was assisting her. No attempt was made to controvert the allegation made in the answering affidavit delivered by Unitas that Mrs van Wyk had all the information which her experts would require in order to advise her. In the circumstances, I respectfully agree with the conclusion of my brother Brand that Mrs van Wyk has not even established the threshold requirement that the report would be of assistance to her.


[53] I cannot, with respect, agree with the conclusion reached by my brother Cameron that Mrs van Wyk has shown that she would enjoy a substantial (and I emphasise the word substantial) advantage, consisting in her being able to particularise her claim and assess its ambit and its viability, if she were to be given access to the report. She is unquestionably able to particularise her claim. The ambit of the claim she sought to pursue in this court appears from the first paragraph of the heads of argument filed by her counsel, which reads (to the extent relevant for present purposes):

From enquiries she [ie Mrs van Wyk] had made and from circumstances she had herself witnessed, [Mrs van Wyk] believes that [Unitas] may be liable to her in damages as her late husband’s death was possibly caused by negligence on the part of [Unitas’] staff giving her a right to claim damages . . . This is the right that [Mrs van Wyk] seeks to enforce . . . .’

In support of this allegation, counsel referred to the following passage in Mrs van Wyk’s replying affidavit:

Everything indicates that there was indeed negligence on the part of the staff of [Unitas]’ (my translation).

As appears from those quotations, it was not argued that it was Mrs van Wyk’s case that Unitas failed in its essential functioning; and Unitas has not had an opportunity to respond to an application for access to the report on this basis. It was by no means clear from her founding affidavit what case she sought to make out. But even if the passage relied upon by my brother Cameron for the conclusion to which he arrives can be interpreted as being the foundation of her case, unless there is a suggestion (and there is not) that the report would establish a causal link between any institutional failure of Unitas and her loss, she does not require the report.


[54] So far as the viability of Mrs van Wyk’s claim is concerned, she did say in her replying affidavit that she wanted to investigate the circumstances of her husband’s death carefully and that she would not embark on litigation if Unitas or its staff were not negligent in causing her husband’s death. But there is no suggestion whatever in the affidavits filed, it was not submitted in argument and there is no reason to suppose that if Mrs van Wyk were to be given access to the report she might not institute action or that Unitas might not oppose it. In other words, there will be no ‘short, sharp end’ to the envisaged litigation that Cameron J held would ensue if the report in Van Niekerk v Pretoria City Council9 were to be disclosed. If access to a record such as the Naude report would merely be relevant to or helpful in the evaluation of a potential claim, but without the prospect that the information (or lack of it) in the record could be decisive, as is the present case, access under PAIA should in my view be denied. To suggest as my brother Cameron does that access should be granted where it ‘can assist’ in avoiding the initiation of litigation, or opposition to it,10 is in my respectful view to give insufficient weight both to the word ‘required’ in s 50(1) and to one of the objects in the statute set out in s 9(b)(i) ie the reasonable protection of privacy and confidentiality.11


[55] For so long as the adversarial system of litigation forms part of our law, issues must be defined in the pleadings and access to documents must ordinarily be limited to discovery after the pleadings have closed. PAIA provides a useful tool where pre-trial discovery of a particular document or documents is required for the exercise or protection of any rights; but the tool must be used carefully with due regard to the facts of each case and the rights of both sides. Resort to it should be the exception rather than the rule. In that way the rights of the defendant, who by definition is a private body entitled to reasonable protection of privacy and commercial confidentiality, as well as the rights of the plaintiff, who before the enactment of PAIA might not have been able to exercise or protect rights properly or adequately, will both be secured.





___________

T D CLOETE

JUDGE OF APPEAL





















CONRADIE JA:


[56] I have had the benefit of reading the judgments of my brethren Cameron, Brand and Cloete. I agree with Brand JA that, generally speaking, whether a record is 'required' depends upon the facts of the case in relation to which the question is being considered.


[57] I therefore think it best not to stray beyond the facts of this case as they appear from the papers. The facts appear mainly, not as one would have expected from the founding affidavit which is nonchalantly terse, but from a statement submitted by the plaintiff to the defendant before the institution of action in support of a demand for the payment of damages, a document that is before us because Unitas Hospital, the defendant, annexed it to its answering affidavit.


[58] Mrs Van Wyk, the plaintiff, says in the statement that her husband, the deceased, was given solid food to eat when on doctor's instructions he was to be fed no solids. That instruction, I infer, had been given because of the deceased's suffering from nausea and continual vomiting while he was being treated in the defendant's high care unit after he had undergone a second operation to excise septic tissue in an infected proctocolostomy wound. While the plaintiff remained with him (which she appears from her account to have done frequently and for extended periods) she could assist the duty nurse in turning to one side the head of the deceased, who was lying flat on his back, so that he might vomit into a bowl: he was unable to turn himself because of the operation wound.


[59] When the night sister came on duty on 1 June 2002 she remarked on the plaintiff's exhausted condition and suggested that she return home, promising that she would look after the deceased well. Shortly after midnight the plaintiff was telephoned to come to the hospital promptly since the deceased had been admitted to the intensive care unit in a critical condition.


[60] The deceased had aspirated which means that he had drawn his own vomit into his lungs. The aspiration incident is admitted by the defendant. We also know that the plaintiff was given the records relating to the nursing care given to the deceased before she launched her application.


[61] As a result of the aspiration the oxygen supply to the deceased's brain was interrupted for sufficiently long to cause brain damage. That is why he was transferred to the intensive care unit where he died about a month later. The plaintiff even annexes a copy of a neurologist's report about the brain damage to her statement. This clearly is what caused the deceased's death and she knew it. There is no suggestion that there was an intervening cause related to the care that the deceased received in the intensive care unit that accelerated his death or impeded his recovery. It would in any event not matter: the legal cause of his death would still be the aspiration.


[62] The only issue in dispute appears to be whether the aspiration was due to negligence on the part of the defendant's staff. The plaintiff is able to demonstrate by her own evidence and presumably that contained in the nursing records as well as that of her medical experts that the deceased who was not be fed solids, who was known to be subject to attacks of vomiting, who was lying flat on his back and could not move himself, choked to death in his own vomit in the defendant's high care unit. If she were to do this, she would in my opinion be doing enough. The defendant would be faced with a prima facie case of negligent treatment. The plaintiff, it is true, mentions in her founding affidavit that conditions were unhygienic, that some of the nurses were not properly trained and that there was an inadequate personnel complement, but she makes no attempt to link these inadequacies to the death of the deceased and with the facts presently at our disposal I do not see how she could.


[63] If the deceased was not given the care that his condition demanded, it is from the plaintiff's perspective unimportant that the neglect to attend to him properly resulted from an institutional shortcoming or some systemic failure. The underlying cause for whatever shortcoming there was in the nursing care would be irrelevant to the plaintiff's cause of action and would not impact upon her right to sue for damages. It could not serve as an excuse for negligence in the circumstances and would therefore not need to be taken into account by the plaintiff in assessing her chances of success in the action or in deciding on the wisdom of compromising her claim.


[64] I agree with Brand and Cloete JJA that the Naudé report, containing as it does, nothing about the deceased's treatment or the cause of his death, would have been of no use to the plaintiff in taking any necessary or prudent pre-trial decision. Since she could comfortably do without it, she cannot be said to have 'required' it within the meaning of that term (as interpreted by this court and others) for the exercise or protection of her right to claim damages.




………………

J H CONRADIE

JUDGE OF APPEAL




Harms JA concurs with Conradie et Cloete JJA

1 Because of the three-year prescriptive time-bar Mrs van Wyk has in fact already had to issue summons, but this is irrelevant since the question before us is whether she was entitled to the report when she initiated these proceedings in February 2004.

2 Interim Constitution s 23: ‘Every person shall have the right of access to all information held by the state or any of its organs at any level of government in so far as such information is required for the exercise or protection of any of his or her rights.’

3 Bill of Rights s 32 (1): ‘Everyone has the right of access to – (a) any information held by the state; and (b) any information that is held by another person and that is required for the exercise or protection of any rights.’ PAIA was the product of s 32(2), which required national legislation to give effect to this right, and empowered the legislation to ‘provide for reasonable measures to alleviate the administrative and financial burden on the state’.

4 2005 (3) SA 486 (SCA) para 13.

5 PAIA s 9: ‘The objects of this Act are – … (e) generally, to promote transparency, accountability and effective governance of all public and private bodies’ (my emphasis).

6 In my respectful view it is the applicant’s exposition of her case in her founding affidavit, and not her preceding letter of demand, still less her counsel’s heads of argument, that should provide the basis for assessing her entitlement to access.

7 See Black v Sumitomo Corporation [2001] EWCA Civ 1819, [2002] 1 WLR 1562 (CA).

8 1997 (3) SA 839 (T) 848G.

9 1997 (2) SA 839 (T).

10 See also the Van Niekerk case at 848F-G.

11 It is interesting to note that in Black and others v Sumitomo Corp and others [2001] EWCA Civ 1819, [2002] 1 WLR 1562 (CA), the case referred to by my colleague Cameron JA, the English Court of Appeal, in interpreting a rule of court which provided that a court may make an order for disclosure before proceedings have started only where (inter alia) such disclosure ‘is desirable in order to . . . assist the dispute to be resolved without proceedings,’ held (in para 81) that for jurisdictional purposes the word ‘desirable’ should be construed as meaning ‘a real prospect in principle’.

▲ To the top