Volume 6, No. 1, Art. 42 – January 2005

Recycling the Evidence: Different Approaches to the Reanalysis of Gerontological Data

Joanna Bornat

Abstract: In 1991 Professor Margot JEFFERYS and two colleagues interviewed "pioneers of geriatric medicine" of whom 60 were geriatricians. These data are now on disk and can be searched digitally. The interviews were long and focussed on the careers of the doctors in terms of their personal successes and their ability to make the undervalued field of elder care into a respectable discipline in academic and practical medicine.

The reanalysis of data for different purposes is an increasingly important methodological issue. This paper considers ethical and methodological issues raised by analysing data generated at another time and by another researcher. Two different approaches, reconstructive oral history and digitised analysis, are discussed with a view to understanding the contribution of overseas trained doctors to the development of the geriatric specialty, mid C20 geriatricians' career choice and experiences of cultural difference.

Keywords: geriaticians, oral history, culture ethics, transcripts, national health service, coding, CAQDAS, Asian doctors

Table of Contents

1. Introduction

2. The Data

3. Ethical Issues

4. Methodological Problems of Secondary Analysis

4.1 Oral history

4.2 Using CAQDAS

5. Conclusion







1. Introduction

In 1991 Margot JEFFERYS and two friends interviewed 73 "pioneers of geriatric medicine" for a project entitled "The Development of the Geriatric Specialty" funded by the Wellcome Foundation. The interviews were long, usually over 1.5 hours, and were transcribed and deposited in the British Library Sound Archive, where the tapes and transcripts can be consulted. Summaries of the transcripts are also readable on the British Library Sound Archive website (http://www.bl.uk/collections/sound-archive/holdings.html#health)1). The secondary analysis of qualitative data is currently a topic much debated in the social sciences and the author of this article is currently, with Dr Gail WILSON, developing further theorisation from the JEFFERYS' interviews. Using a reconstructive oral history and digitised approaches it discusses such topics as the contribution of overseas doctors to the development of the geriatric specialty, luck and chance in geriatrician career choice and doctors' encounters with cultural difference. In doing so it also brings into view a particularly rich set of data which has continuing significance for the health and social care of frail older people fifteen years after its origination. Grand aims, however in undertaking this project it rapidly became clear that reusing data is far from simple. There are ethical and methodological problems which are explored in what follows. [1]

2. The Data

Margot JEFFERYS chose a theoretical sample of the people who she believed could tell her most about the development of geriatrics. As far as it can be seen she had a straightforward historical motivation to capture the experience of the "pioneers" before they died. The oldest was 92 in 1991 and 18 had been born before 1914. The theoretical model behind her approach appears to be that change results from the activities of key individuals. Though it seems unlikely that Margot JEFFERYS would have seen this theory as adequate for a full study, it is clearly embodied in the interviews. The theoretical sample was mainly doctors (geriatricians, with a few psychogeriatricians and general practitioners) but included pioneering nurses, social workers, an occupational therapist, civil servants and officials from voluntary organisations, and two Conservative ministers, Kenneth ROBINSON and Enoch POWELL (JEFFERYS 2000). [2]

The interviews can be described as guided life histories (THOMPSON 2000, ch.7). Interviewees were asked about family and early education in so far as it accounted for their going into medicine, but the main focus was their careers and their views on how geriatrics developed. [3]

There are physical problems with the data. The transcripts can only be accessed under controlled conditions at the British Library. This may be adequate for some forms of historical research but not when digitisation is planned. The transcripts can be photocopied for a price but only by the British Library because of the risk that outsiders will damage the originals. Many of the resulting photocopies were not very legible. The combination of early word-processing and less than perfect photo copying made scanning the documents very difficult. Each had to be checked against a text that was itself often difficult to read. It would of course have been possible to go back to the originals which were probably all legible but the time taken was already very great. We were confident, however that few errors (maybe none) were introduced in the checking stage. Once scanned and cleaned, the data could be put into a form suitable for computer assisted qualitative data analysis, thus opening up a whole new set of ethical and methodological problems. [4]

3. Ethical Issues

Participants gave their interviews in good faith to help illustrate the history of geriatric medicine. They placed their interviews in the public record. Margot JEFFERYS checked back with some of them on what they had said by sending a transcript of the interview, but we do not know who, nor how many. Since the time of her interviews, sending back transcripts has become common and indeed expected practice amongst oral historians at least. [5]

Their personal narratives were given in good faith, but what does it mean if the interviews are now used to look at, for example, the culture of practitioners of geriatric medicine, and if a new set of interpretations is introduced? Would the respondents have agreed? We can be certain that some would have said no. Only 11 out of the 54 doctors' transcripts when searched on disk used the word "culture" or a version of it such as "cultural" in a social science sense. The interviewees were not social scientists and the word was not in their vocabularies. It seems possible that some, if not many, would have been hostile to the problematisation of their world views that is implied in analysis of cultural difference. The question however is not so much whether respondents used the word culture, but whether they saw cultural difference as "natural" (the working classes are not quite like us) or as a problem. And where it was a problem (like non compliance or poverty) what they thought about it. [6]

We have no way of knowing who would have refused to participate. Not all the original interviewees are dead, however, the terms of their agreement to deposit allow for their words to be on the public record. Two issues arise, both relate to the issue of control over data when it is deposited. First, there is the question of whether new use of the data could be seen as introducing a form of deception when gaining informed consent, if an interpretation identifies a focus not introduced at the time of the interview. This problem is likely to apply to the reuse of all qualitative data sets, in fact to any data at all. Indeed it may not even require time to elapse for varied interpretations of data to emerge. Readings of the same transcript by different people can yield quite different interpretations. As this is often an accepted part of data analysis during the lifetime of a project it would be overly rigid to exclude this practice at a later date (BORNAT 2002). Indeed the lapse in time highlights the unreality of informed consent, even when participants are given the most careful and detailed explanation, as Janet HEATON suggests in her review of practice in secondary analysis (HEATON, 2004, p.80). In on-going projects, the more value laden the data the more possible it is for participants to feel their words have been twisted or misunderstood by analysts with different values from themselves. When period effects of changes in intellectual approach are combined with different disciplinary standpoints (medicine and medical sociology or oral history and social policy) the implausibility of informed consent become clear. However, as Wendy HOLLWAY and Tony JEFFERSON have also argued, there may in the end be "a limit to the issue of consent" (2000, pp.94-95) if we are also to adhere to principles of freedom of speech (BORNAT 2002, p.52) valuing critical investigation as the way that "intellectual understanding advances" (CORTI & THOMPSON 2004, p.337). [7]

The second issue relates to anonymisation. Qualidata argues that "The objective should be to achieve a reasonable level of anonymisation" (their emphasis) when preparing data for archive deposit (http://www.esds.ac.uk/aandp/create/identguideline.asp; accessed 06.01.05). However, even if quotations are anonymised for the purposes of writing on culture or any other possibly controversial topic, once the data is changed from being paper based to online, they are searchable and quotations may be identified. If, for most social science research, large bodies of qualitative data need to be digitalised in order to be useful to secondary analysts, this problem will arise whenever data has not been anonymised very carefully. Simply changing names, even place names, will often no longer be enough. [8]

There is, however, another issue. In the case of the geriatricians the anonymisation of the original data would have been inappropriate. The aim of the interviews was to record the personal contributions of the pioneers and their colleagues as seen from their own perspectives. Without details of their careers, as medical students, young doctors and pioneers in their field, the data loses its value. And, the fact that they were willing to place their interviews in the public view indicates that they were willing to be identified. The anthropologist Barbara MYERHOFF in Number our Days, her study of elderly Jewish members in a senior centre in California found a similar situation where her informants were insulted by the idea of anonymisation. They wanted their contributions and their lives to be placed on record. They wanted to be "stars" in her film and thus named (MYERHOFF 1978). However it can be argued that these wishes assume that the wider public will evaluate participants contributions in much the same way as they do themselves, that they and their texts will be treated with respect. In the end, for both sides, the law is perhaps the best guide to practice as the Oral History Society's web pages make clear (http://www.oralhistory.org.uk/ethics/; accessed 05.01.05). [9]

Before I go on discuss methodological issues for secondary analysis I will look briefly at some further ethical issues: those of changed language use, intervening in an interview relationship, and not "being there". [10]

In terms of language, today's controversial issues were spoken of quite differently in former times, sensitivities are differently minded, questions of class, race and gender are raised in ways which previously would not be recognised. In some respects, these have been rapid, yet uneven changes (GUNARATNAM, 2003, pp.11-18). The result is that we are faced with the question of how to deal with changed language use when returning to data. In an earlier paper I discussed the ethics of introducing an alternative analysis of the JEFFERYS' data based on references to race and ethnicity (BORNAT, 2002. The specialty of geriatrics was, it seems, dependent on recruits from the Indian sub-continent if it was to expand. Several of Margot JEFFERYS' interviewees were quite explicit about this:

"One of the problems has been that staffing of geriatric departments hasn't always been easy, we have had to appoint quite a lot of doctors from the Indian sub-continent to be registrars and even senior registrars, so for quite a period the only applicants for consultant jobs were in fact not British citizens trained by British methods."

"They had been to respectable geriatric departments and learnt the trade but when they got appointed to x, y, z, they had Indian or Pakistani names or whatever else. And it tended to get known as the sort of, you know, dark-skinned specialty. Third world specialty almost." (AGATE, British Library catalogue, C512/8/01-02). [11]

Of course there are benefits to comparing forms of expression and language, across time. We can mark change and also reveal continuities in expression which we might not be aware of. Dealing with what may now be regarded as unacceptable language can be a challenge, but it is not confined to revisiting older data. What we might need to acknowledge is that the interviewees were recalling a time when racial difference, if not racial superiority, was unproblematised at a time, the early 1990s, when identification of racial difference as in any way "natural" was becoming largely unacceptable. The question of how to interpret language is therefore complex in the extreme and points to a need to be aware of the interplay of context when attempting to explain or identify cultural difference (SAID 1993, p.234). [12]

Oral historians often argue that their approach to interviewing, with its enabling and empowering qualities, its sensitivities and intimacy, leads to a personal relationship, one which endures beyond the interview (RICKARD 1998; ROUVEROL 2000; THOMPSON 2000; BURTON 2003; BAKER 2004). Some sign up to a "shared authority" in the production of the transcript or publication (FRISCH 1990; SITZIA & THICKETT 2002). Secondary data analysis must inevitably lead to intrusion in such circumstances. Over time of course interviewers, and interviewees too, may distance themselves in various ways from the experience. The act of depositing in an archive is a way of involving a wider audience, recognising that others might have an interest and a share in the ownership of what has been discussed. As far as the geriatricians and Margot JEFFERYS and her colleagues were concerned, the aim of the project was to ensure a place in history for a marginalised specialty. Even if the intimacy of the interview does not always survive it may have an effect on what is discussed and how the dynamics of an interview develop. Not "being there" reduces the interview to the text alone and changes the way the text can be analysed (HEATON 2004, pp.60-61). The interview texts become decontextualised so the secondary analyst is working with less data. This may allow a more "objective" approach but it excludes all the additional data on tone, emotion, expression and body language that an original interviewer can remember assuming that interviewer is also the data interpreter, unless these are included in the transcript and supporting notes (which they are not in this case) (THOMPSON & CORTI 2004, pp.336-337). Any theoretical assumptions informing the interview process can be identified fairly straightforwardly as I suggest later. Similarly, the historical and social context of the interviews can be identified without too much difficulty, however, the more intimate and inter-personal exchanges may be more elusive, though they may provide endless opportunity for speculation and reflective interpretation. I come back to these points later. [13]

4. Methodological Problems of Secondary Analysis

Given that the original interviews were life histories a follow-up using a reconstructive oral history approach would seem to be appropriate (THOMPSON 2000, pp.286ff.). [14]

4.1 Oral history

The 54 interviews with the doctors in the sample, as I have already pointed out, reveal certain shared characteristics. They comprise two cohorts, those starting their professional careers before the National Health Service (NHS) became law in 1946 and those who qualified after it was set up in 1948 (WEBSTER 2002). Some had pre-NHS medical experience, the majority did not. All except five were men and all but one was white, brought up either in the UK or Europe. Using a reconstructive oral history approach we can probe further, seeking out evidence for those social and economic factors which led to the career of geriatrician for this group, looking for expressed themes and noting interview interactions, in such a way as to determine the cultures which characterised them. [15]

Some of the characteristics of the geriatricians are identifiable simply by counting and of course it is possible to build on this by reading through the transcripts and noting just how many came from families where there was already a doctor, or from schools which provided a direct route into university. We might also identify which medical schools they attended, where they worked and which other geriatricians' names they mention. From this it is possible to build a picture of a group of doctors who were middle class, often came from families where there was already a medical connection and who could readily identify the most helpful features of the medical network. [16]

However, further reading through indicates a rather different picture, one that suggests that, for the older interviewees at least, for whom there was no geriatric specialty until they created it, their medical career was marked by setbacks as well as opportunity, personal, social, and historical. Many of these doctors' careers were interrupted and only partially reinstated by wartime service, others were not fortunate when it came to career opportunities, yet others again saw in geriatrics a chance to progress rapidly and to innovate medically and administratively. The advantage of drawing on a whole life, as described in an oral history interview, is that experiences such as war service, the founding of the NHS, and provision for frail older people in pre-NHS days are revealed in all their structural complexity. As for example, one founder who describes a formative incident in his first job as a registrar at Fulham hospital after an education at Cambridge University, Guys Medical School, war service in the Far East and consequent late demobilisation:

"And she so she said, 'Now I've got these prescription forms for you to sign' and do you know, I was just about to do it and I said, 'Wait a minute, sister, I have never in my life signed a prescription for a patient whom I have not seen and I'm not going to start this afternoon. I will see all these patients before I sign these prescriptions'. And the sister nearly fainted at that point because this was the first time any doctor had insisted on seeing the patients. Well, I was shocked, amazed, appalled, saddened and very upset at what I saw that afternoon. I could not believe that, within the campus of one hospital, two separate standards should exist for treatment based solely upon age, 65 for men and 60 for women." (VINE, British Library catalogue C512/68/01-02). [17]

He goes on to identify this moment as a turning point, leading him to challenge the exclusion of patients deemed as old from acute services. In this one life story, as in the others, there is evidence of the structural determinants, of class, politics and work-based cultures which shaped medical careers as well as the agency of these first geriatricians when it came to challenging organisational and professional structures. Along with other interviews in this collection we are given insights into medical and hospital cultures and the behaviours, expectations and relationships which support them. [18]

Identifying accounts and statements which provide evidence to be weighed, compared and at times challenged is a more or less straightforward oral history method. Searching for expressed themes, patterns and forms of expression which could provide evidence of perspectives, meaning, reflection and evaluation of a life is a complement to this approach. A read through the data suggests some interesting recurrences. There are references to luck and chance in many of the transcripts, as for example:

"So I've been very lucky in the environment in which I've worked" (COHEN, British Library Catalogue, C512/23/01).

"I was lucky, I got scholarships which paid for maybe four-fifths of my medical training" (GODBER, British Library Catalogue, C512/28/01).

"I was lucky inasmuch they were terribly thankful to have me" (MORTON, British Library Catalogue, C512/4/01-02).

"It was pure chance that it has become so interesting" (FINE, British Library Catalogue, C512/17/01-02).

"But I still continued, I didn't abandon the geriatrics, as some people have done once they got a chance. I felt that—I may be conceited—but I felt that I was doing a good job and that I had something to give them" (BENIANS, British Library Catalogue, C512/55/01-02).

"A lot of people make a very good living or a good name by the sheer chance of being tied in with a process which was going to happen anyway, and that's what happened to me" (ADAMS, British Library Catalogue, C512/19/01-02). [19]

Elsewhere I have looked at the significance of the theme of luck and chance in the story of geriatricians, concluding that they suggest uncertainty and ambivalence, in accounting for career histories and achievement in a low status branch of medicine (BORNAT 2004). Within these chronologically determined accounts, more deeply embedded aspects of the culture of geriatric medicine are expressed. [20]

Finally what can be learned from interview interactions? The interrogative nature of oral history injects a dynamism into accounting for the past which enables people to question and reformulate awareness of their lives (PORTELLI 1991, p.258). Being interviewed can be a process of becoming strange to oneself, as much as an opportunity to perform and express feelings in response to questions which, in the asking, may have unpredictable consequences. In the Margot JEFFERYS' interviews there are factors which might have mitigated against the unexpected. She was already well acquainted with several of the interviewees and in some cases it is clear that the tone of the exchange is affected by an assumed and presumed elite status on the part of the doctor being questioned, which if it did not permeate Margot JEFFERYS' interviews certainly comes through in those of her co-interviewers. [21]

Identifying structures, themes and interactions are all part of a reconstructive oral history approach to data interpretation and analysis. The whole life story, its positioning within historical time and social and economic determinants, provides a key to understanding processes and relationships as well as cultural expressions and subjective meaning. However computer assisted qualitative data analysis (CAQDAS) allows these understandings to be examined in greater depth and can shorten the time needed to substantiate original interpretations. [22]

4.2 Using CAQDAS

One of the aims of the secondary analysis will be to consider how differences in culture between "Asian" geriatricians (those trained in India, Pakistan, Bangladesh and Sri Lanka and working in the NHS) and the majority of their patients, affected practice and attitudes to patients. In order to consider culture as more than an exotic attribute it is helpful to consider differences in cultures between white (all but one of the original interviewees) geriatricians and their patients. [23]

As I've already pointed out this means starting with a theoretical sample that was drawn for a different purpose. This is not necessarily a disaster if the aim is to consider how cultural difference did or did not affect the attitudes of the pioneers of geriatric medicine, as opposed to selecting a theoretical sample to study the effects of cultural difference on geriatric medicine. There is no British work that we have been able to find on cultural differences of doctors from their patients, as opposed to the cultural differences of patients which have a large literature centring round non compliance and poor understanding of communications (SHAHEEN 1991; SINGH 1994, ESMAIL & CARNALL 1970). At this early stage in the development of knowledge therefore, the difference between the two types of theoretical sample appears to be unimportant. [24]

I want to argue that it is only possible to look for a complex concept such as cultural difference in large data sets if they are digitalised and even then it is not easy. The "correct" method is to proceed through each interview coding for "culture" or for various aspects of culture. This implies that the definitions of the term and its variations and sub codes are known in advance. In other words the research is not using grounded theory but is imposing a set of preconceived definitions on the data. This process is very much easier in primary data analysis since the concepts implied in the coding frame, even if imposed are likely to have been developed by the original researcher either before or during data collection. In secondary analysis where the data is much less familiar to the researchers and the concepts being used were not envisaged at all in the original research design, the analysis is necessarily less secure. Coding schemes may be devised, either by building from the text or developed from the values of the researchers, but either approach will be difficult if the data set is large and varied (HEATON, 2004, pp.96ff.). Decontextualisation (see below), is an even greater hazard. CAQDAS does not solve these problems but it enables a more thorough approach to analysis and should produce a more trustworthy set of conclusions. [25]

As mentioned earlier, there were only eleven doctors who mentioned culture in a social science rather than a natural science context. They were either talking about their work overseas, or discussing the changing culture of the NHS, so none of these mentions is useful for the present analysis. [26]

Early reading of the text showed how shocked many young doctors were by the conditions they found in the old workhouses (officially abolished in 1929 but living on as warehouses for the old and poor and as casual lodgings).This threw up questions of difference. Coding established that none of the doctors interviewed appeared to have a manual background. Two talked so persistently about their professional lives and their views of geriatrics (and geriatricians) that the interviewer left without asking about their family back grounds. Two were the sons of "Ulster farmers" not farm labourers, so they too were coded as middle class. [27]

One other born in 1911 only says "my people were divorced". This is English public school language, but was it acquired early or late? Virtually all others clearly state that their fathers were white collar if not higher professionals. Since geriatric medicine appears to have been almost exclusively based in former workhouses pre 1948 and mainly so for at least another twenty years after that (OGG, EVANS, JEFFERYS & MacMAHON 1999), cultural differences relating to class definitely existed:

"... I mean walking into chronic wards in the workhouses in this country in the 1940s and '50s you practically needed a clothes peg on your nose. You could tell going through the door what the rate of incontinence was, or what the lack of staff was because the fewer the staff and the greater the difficulty they worked under, the greater the stench, it's as simple as that." (British Library Catalogue, Adams, C512/19/01-02) [28]

The same class difference existed for the rest of the medical profession, outside the more prestigious hospitals, but workhouse hospitals housed the "old, infirm and immoral" so the difference was more stark:

"And the VD (venereal disease) department [of the workhouse] was part of the dermatological section, of course. The other doctor was a lady, and she looked after the female VDs, and I looked after the male VDs. It may surprise you, but male VD is a lot more wholesome to deal with than female VDs. When Dr Peacock was on holiday I had to do the female VDs, and equally she did the males when I was away. And really, I was horrified, first of all, by the coarseness of the women, the young women there, and their lack of any sort of reticence about their problems down below. Oh dear, it put me off for a long time, having to do female VDs." (British Library Catalogue, Nagley, C512/58/01)


"The sights one saw there were really pretty pathetic. I'll never forget, just having been in the navy myself, there was one character there who was about 80. He was grossly overweight, he was incredibly coarse, and he would wander up and down the long corridor supporting himself on a wheelchair, shouting at the top of his voice. And I would usually say. 'All right, Chiefie, is it tot time?' And he said, 'Boy, I've pissed better rum than they get in the navy these days', and he'd go off roaring with laughter. And that was the sort of atmosphere of the workhouse there." (British Library Catalogue, Morton, C512/4/01-02) [29]

Cultural differences therefore existed but how do we extend the evidence from the stark cases quoted above and how do we show that they affected practice or outcomes? The first methodological problem arising from any computer assisted qualitative data analysis is decontextualisation. In this project that is a double problem. The focus of the original interviews was on the participants as individual heroes of medical innovation. They fitted well into the established view of the professional as functioning in relation to "his" patients rather than as part of a system or organisation. As geriatricians taking a holistic view of patients and often learning to negotiate with "powers that be", they mention social and organisational issues, but usually more as hindrances to professional activity than seeing themselves as integral parts of a system. The interviews therefore appear very decontextualised to a secondary analyst. For the original interviewers however they were not. Margot JEFFERYS had been a founder member of the Keppler Society, the forerunner of the British Geriatrics Society, and as a medical sociologist she personally knew many of the interviewees, one was her own son and she had even employed another on an earlier research project. One of her co-interviewers was a retired social worker with wide experience. For them the interviews were in context. Her team also collected articles and other publications that were relevant from the interviewees, and followed up references they were given. In a modern project these would be clearly listed as off line documents but in the 1991 there was no way of doing this and they are lost. So we come to a set of de facto decontextualised interviews and proceed to use CAQDAS to analyse difficult concepts. Full transcription combined with a programme that allows all text searches to be "spread" can partly overcome this problem. Spreading places the chosen coded item within the context of the interview so that what went before and what followed is available and can be added to the coding, or eliminated as needed, to make sense of the coded item. However this only enables coded text to be placed within the overall interview, it does not solve the problem of decontextualisation of the interview itself. All secondary analysis is therefore likely to be more insecure than primary analysis. [30]

Two approaches to the problem are possible both using CAQDAS (QSR N6 in this case see http://www.scolari.co.uk/frame.html?http://www.scolari.co.uk/qsr/qsr_n6.htm; accessed 12.01.05) but others work as well. The first is to search the text for likely words or phrases. The question here is what do you lose and what do you gain by the failure of words to name concepts. (Put explicitly if ironically: the failure of interviewees to use the vocabulary of social theory and their failure to see social theory as relevant to their lives and world view). A search for "patients" turns up far too much data for a preliminary approach but will be useful for Boolean searches at a later stage when overlaps between the text for patients and texts for other aspects of behaviour or belief can be produced. More successfully a few key documents can be fully coded and then text searches can be iterated, guided by what is found, in order to extend the evidence. The aim is to investigate the range and typologies of cultural differences rather than to establish how widespread their existence and effects may have been. Since this is a theoretical sample there is no point in calculating averages or any other form of statistical analysis but it is helpful to know how much data are likely to come from individual searches. As noted above the workhouse was a key landmark in early geriatrics and an indicator of extreme cultural difference. Hence a search for "workhouse" shows that 41 out of 54 interviewees mentioned it in some context or another. The assumption is that what they say about the workhouse will often give some idea of how they felt about their older patients. And sometimes it does. Other mentions are purely factual but provide context and background. Since so many early geriatric wards were in workhouses, that fact may be all that is mentioned. [31]

The next stage, for the research which Gail WILSON and I will be carrying out is to find other indicators of cultural difference and to build up a network of codes and sub codes grounded in the data. We hope that these results will be published in a later paper. [32]

5. Conclusion

The Margot JEFFERYS' interviews are lodged in the British Library where they are open to be consulted by anyone with an interest in the wide range of topics which the data set includes. In discussing an interest to attempt a secondary analysis of her research we have a number of motives of our own. [33]

First it is our conviction that these data are extremely rich. They provide a detailed, if at times partisan, account of the emergence and development of the NHS and of the medical care of older people between 1935 and 1990. Margot JEFFERYS' original intention in interviewing these pioneers may have been simply to identify their contribution to the founding of the specialty, however what she created is incomparable detailed and extensive evidence and in returning to the data she created with her colleagues we are acknowledging her achievement. [34]

To reuse data, whether one's own or another's, is to return with new perceptions, possibly better informed and certainly with the benefit of additional contributory data sources. Our initial interest was sparked when we noticed several references to overseas trained doctors and we have pursued that with subsequent interviews and research. As we have suggested, to reuse data is to introduce some difficult ethical issues, both in relation to consent and reinterpretation. Though challenging, we feel that these issues can be resolved and indeed must be if research which draws on individuals' views and experiences is to continue. [35]

Our final motive is one which we feel must be self-evident from the data which cited in the paper. As gerontologists we should be taking heed of the accounts these pioneers give of their original struggles and later achievements. They were describing a root and branch ageist treatment of older people and they were seeking ways to redress this inequity and discrimination. At the time of their interviews, 1990, they were witnessing the dismantling of what they had achieved. [36]

Several expressed extreme reservations about the coming privatisation of medical care for older people. Few at all had taken up private practice, most had what appears to be high, if critical, regard for the NHS. In founding the geriatric specialty, those who worked in England (the Scottish system was different) were seeking to include older people within the best acute and continuing care services of the general hospitals. As others have argued, the policies of foundation hospitals together with the co-production of health and social care provision takes us back to segregated and unequal provision (POLLOCK 2004). To read the interviews is to become aware of the retrogressive nature of such policies and to be extremely concerned for the present and future care of frail older people. [37]


I would like to thank Dr Gail WILSON, Emeritus Reader, London School of Economics for her generous contribution to the writing of this article.


1) Rob PERKS, oral history curator at the British Library wrote, in answer to an enquiry, "The entire collection (C512) has summaries on the catalogue, except for those that are closed to public access. A search under 'C512' in the online catalogue at http://www.cadensa.bl.uk/ finds 147 records: the 'Recording' entries provide the summary. 73 interviews were deposited of which 70 had summaries. Ten are closed. I have a note from a meeting with Margot and Hazel Houghton on 21 January 1993, when they were about to deposit the collection, that says 'Transcripts for all interviews: some corrected by interviewees'". <back>


Baker, Lorna (2004). Making money and memories: running an oral history business. Oral History, 32(1), 79-86.

Bornat, Joanna (2002). A second take: Revisiting interviews with a different purpose. Oral History, 31(1), 47-53.

Bornat, Joanna (2004). Chance as narrative theme or pragmatic function?: geriatricians recall their careers. Berlin: European Social Science History Conference.

Burton, Sarah, K. (2003). Issues in cross-cultural interviewing: Japanese women in England. Oral History, 31(1), 38-46.

Corti, Louise & Thompson, Paul (2004). Secondary analysis of archived data. In Clive Seale, Giampetro Gobo, Jaber F. Gubrium & David Silverman (Eds.), Qualitative Research Practice (pp. 328-343). London: Sage.

Esmail, Anwar & Carnall, Douglas G. (1997). Tackling racism in the NHS. British Medical Journal, 314, 618.

Frisch, Michael (1990). A shared authority: essays on the craft and meaning of oral and public history. Albany: State University of New York.

Gunaratnam, Yasmin (2003). Researching "Race" and Ethnicity: Methods, Knowledge, Power. London: Sage.

Heaton, Janet (2004). Reworking Qualitative Data. London: Sage.

Jefferys, Margot (2000). Recollections of the pioneers of the geriatric medicine specialty. In Joanna Bornat, Rob Perks, Paul Thompson, & Jan Walmsley (Eds.), Oral History, Health and Welfare (pp.75-97). London: Routledge.

Myerhoff, Barbara (1978). Number Our Days. New York: Dutton.

Ogg, Jim; Grimley, Evans J.; Jefferys, Margot, & MacMahon, Douglas G.. (1999). Professional responses to the challenge of old age. In Bernard, M & Phillips (Eds.), The Social Policy of Old Age: Moving into the 21st century (pp.112-117). London: Centre for Policy on Ageing.

Pollock, Allyson M. (2004). NHS plc: The Privatisation of our Health Care. Verso: London.

Portelli, Alessandro (1991). The Death of Luigi Trastulli and other stories: form and meaning in oral history. Albany: State University of New York.

Rickard, Wendy (1996). Oral history—"More dangerous than therapy?" Interviewees' reflections on recording traumatic or taboo issues. Oral History, 26(2), 34-48.

Roberts, Celia; Sangari, Sarangi; Southgate, Lesley; Wakeford, Richard; Wass, V. & May, Carl (2000). Oral examinations—equal opportunities, ethnicity and fairness in the MRCGP. British Medical Journal, 309, 320-375.

Rouverol, Alicia (2000). "I was content and not content": oral history and the collaborative process. Oral History, 28(2), 66-78.

Said, Edward (1993). Culture and Imperialism. London: Chatto & Windus.

Shaheen, M.Z. (1991). Problems of new overseas doctors in the United Kingdom. British Medical Journal, 302, 966.

Singh, Swaran (1994). Cultural adjustment and the oveseas trainee. British Medical Journal, 308, 1169.

Sitzia, Lorraine & Thickett, Arthur (2002). Seeking the Enemy. London: Working Press.

Thompson, Paul (2000). The Voice of the Past (3rd edition). Oxford: Oxford University Press.

Webster, Charles (2002). The National Health Service: A Political History (2nd edition). Oxford: Oxford University Press.


Joanna BORNAT is Professor of Oral History in the Faculty of Health and Social Care at the Open University, UK. She has researched and written on aspects of remembering and of later life and has recently been looking at the secondary analysis of archived qualitative data.


Joanna Bornat

Faculty of Health and Social Care
The Open University
Walton Hall, Milton Keynes, MK7 6AA

E-mail: j.bornat@open.ac.uk


Bornat, Joanna (2005). Recycling the Evidence: Different Approaches to the Reanalysis of Gerontological Data [37 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(1), Art. 42, http://nbn-resolving.de/urn:nbn:de:0114-fqs0501424.

Revised 6/2008

Copyright (c) 2005 Joanna Bornat

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.