Sue Ziebland
Keynote Lecture
KL 5 / How to include patients’ perspectives in the development of health promotion strategies for primary care

Abstract

There are political, practical and economic reasons why the consumer perspective on health promotion is considered important. However, there are still debates about how users’ views should be included. Despite some emerging examples of excellent practice, the logistics of identifying and involving consumer representatives have sometimes been an obstacle. This paper discusses some of the ways in which patients' perspectives are sought, and considers the potential contribution of research that is designed to identify lay perspectives. Qualitative research methods, which include observations, individual in-depth interviews and group interviews, are designed to explore the beliefs and perceptions of the participants. Some qualitative researchers are making efforts to collect, analyse and present their findings so that they can be used to help inform patient groups, health professionals, researchers and policy makers about the issues that are important from the patients' perspective. Such work can be used to support or supplement the participation of consumer representatives on committees and advisory groups. This paper discusses how qualitative research, and a new project called DIPEx (database of individual patients' experience) could be used to form health promotion strategies.

Background

Why the consumers’ perspective is important.

The importance of involving patients, or consumers, in the shaping and delivery of UK health services and health care research has been stressed in the NHS Executive Patient Partnership Strategy (Department of Health 1996) and the need to recognise the potential contribution of expert patients was highlighted in ‘Saving Lives: Our Healthier Nation’ . Patients are no longer viewed as passive recipients of health care, accepting that the doctor knows best, and are increasingly characterised as critical consumers(1). This suggests a quite remarkable change in the power relations between patients and doctors, although it is likely that the reality of many clinical encounters lags behind the rhetoric. Reasons why decisions are not always discussed include poor communication between patients and health professionals and reluctance, on the part of one or the other, to acknowledge that there is uncertainty about the most desirable or effective treatment. Furthermore, not all users are willing or prepared to take a full part in making decisions about their health care and, in the UK, those who are in the greatest need are least likely to be involved in this way. However, encouraged by the popular media, the Internet, changes in medical education and the expectations of health professionals, awareness of different treatment options and expectations about the role of patients in decision making is changing . Mechanisms are needed, and are being developed, to elicit patients' perspectives that are neither filtered through a clinical prism(2) nor dismissed as unrepresentative or subjective.

When health promotion interventions are being designed, there are practical and functional reasons why the patients' perspective should be sought. If approaches are not developed with users and other relevant groups, the content or style of the intervention may be inappropriate. At best, the intervention might be ineffective, at worst it could jeopardise the relationship between the patient and the health professional(3)(4). For example, midwives may be reluctant to (offer repeated advice to) repeatedly advise their clients to quit smoking out of concern that it may damage their rapport and prevent other important contacts during pregnancy. People who are overweight and feel that they cannot visit their GP for any purpose without their weight being mentioned, may become unwilling to seek help for their health problems. Interventions that aim to change health behaviours are more likely to be effective if the mode of delivery, style and balance of familiar and novel information is developed through research and consultation with the target groups(5).

Decisions to change behaviour are complex, but people who feel that their views and preferences have been taken into account may be more inclined to try to change damaging health behaviours. However, this relationship is not straightforward. Conclusive evidence about the effects of patient involvement in decision making is difficult to provide, not least because there is debate about what would be the most desirable result. Candidate outcomes include the patient's satisfaction with the decision, changes in their knowledge about the options, and awareness of uncertainty, as well as the more conventional outcomes associated with improvement in health status. If these variables all point in the same direction it is unlikely that there would be a dispute about the positive effects. However, a well informed patient with hypertension can weigh up the pros and cons of trying to lose weight or quit smoking and conclude that, on balance, they would rather not make the effort. This is illustrated by the following example from a hypertension study for the DIPEx project:

I think that if I could lose weight, if I could eat less salt … The last thing I am likely to do is give up alcohol, which I like very much, and I find that very difficult to give up and I do like smoking the odd cigar as well. So I put myself in a high risk category .. I know everything I should be doing. But life is for living .. to give up all those would be bloody miserable. I would rather titrate that against my hypertension. RM15.

Another reason why the patients' perspective is increasingly being seen as important is the realisation that there are considerable financial penalties to the misunderstandings between patients and doctors. Concordance studies have noted that poor communication in the consultation often results in wasted medicines, as well as alienation between the doctor and patient. Doctors feel that patients are demanding prescriptions while patients contend that unwanted medications are being thrust upon them. Research that combines clinical observations with interviews before and after the consultation, with the aim of understanding the perspectives of clinicians and patients, has begun to identify why these misunderstandings persist(6). The savings that could be made on the national drugs budget may be persuasive for any who remain unconvinced by other arguments about the importance of eliciting the patients' perspective. However, although several studies have indicated that increased involvement in decision making may produce less demand for medical interventions, there is no guarantee that greater patient participation will result in overall cost reductions(7).

Methods of involving the consumers' perspective

Lay membership of advisory groups

The most frequently discussed method of involving consumers is as individual members of advisory groups, steering groups or members of consumer networks, such as the Cochrane collaboration consumer groups. Many, but not all, voluntary groups have strong user representation on their policy and funding committees. The proportion and influence of the lay and professional members on these groups varies considerably, but the existence of the NHS advisory group on consumers' involvement in health care and several years experience can help to guide groups with respect to best practice. An example is the UK Alzheimer's Society which has established a network of supporters (mainly informal carers or ex-carers of people with dementia) who are willing to act as consumer monitors for research projects funded by the Society. Involvement at an early stage of project development can mean that projects are shaped to address issues that are of concern to the user group, rather than merely to react to the research results. Similarly, the Cochrane collaboration encourages consumer involvement in shaping the topics and questions for systematic review, as well as comments on the results.

In a recent BMJ editorial, Williamson presented guidance for involving lay members on doctor-patient groups formed to discuss 'standards and guidelines, audit and revalidation'(8). Among her recommendations, she stressed the importance of ensuring an even numerical split between lay and professional members and of encouraging a low turnover of group members, since initial barriers will be broken down if members learn to respect each others expertise. She also emphasises that clinical relationships between the group members should be avoided and that members' contributions should be valued as their personal experiences, rather than for the extent to which they are believed to be 'representative' of a patient group. While the atypical characteristics of lay members are sometimes cited as a reason for dismissing their contribution, this position suggests a double standard. The professional members of a group will usually be invited to join because of their particular, rather than typical or representative, interest in the area. It is inconsistent to value professionals for their unusually strong interest in the field, while 'discounting' the views of lay members as unrepresentative.

Qualitative research with consumers

Another approach to identifying consumer priorities, albeit one which has been less frequently discussed in these terms, is via qualitative research, which uses methods that are specifically designed to identify lay perspectives(3). Qualitative research methods, which include observations, individual in-depth interviews and group interviews, explore the beliefs and perceptions of the participants, within their natural setting. This can mean observing behaviour in its social context, or exploring issues in interviews that are led by the priorities of the participant. The potential contribution that qualitative research methods could make to informing the development of health promotion strategies is discussed below.

Consumer perspectives can be explored using a range of methods including surveys, focus groups and individual interviews. However, not all are of equivalent value in exploring patients' perspectives. A detailed discussion is beyond the scope of this paper but broadly, where the issues are already reasonably well understood, perhaps from earlier qualitative work, and the aim is to establish the numerical frequency of a view or experience within a given population, a self-completion postal questionnaire may be most appropriate. This approach has the advantage of reaching a large and representative sample, however it unlikely to uncover new issues, since the questionnaires necessarily explore known variables. Open-ended questions can be included to try to identify unanticipated issues, but since not all respondents are willing to complete these sections, there is a danger that the responses will only represent the views of those who are more literate.

Focus group interviews in which issues are discussed between several respondents can sometimes identify new and unanticipated aspects of peoples' experiences. In some cases, participants may be emboldened and encouraged to discuss sensitive issues that they would be reluctant to talk about individually(9). However, group interviews may produce shared accounts that are more limited in range and more conservative than the accounts derived from individual interviews. Focus group researchers are usually advised to ensure that the groups are homogenous in terms of age, gender and social class and that an experienced, trained facilitator encourages all respondents to express their views. It is very important that group dynamics are not ignored in the analysis of focus group data and the data should certainly not be analysed as if equivalent to the sum of individual interviews. Although group interviews may initially seem appealing to a time-limited research team, unless the aim of the research is to uncover shared beliefs or identify issues of contention, it may be easier and more productive to analyse individual interviews.

Semi-structured telephone interviews allow the researchers to study a larger population, with a less time-consuming design than is possible with face to face interviews. If a numerically representative sample is required, or the population is geographically dispersed, yet it is desirable to include open ended questions, a telephone interview may be worth considering. There are limitations with the approach, particularly if the subject of the interview requires a lengthy discussion or a degree of rapport that is unlikely to be achieved over the telephone.

Face to face, in-depth individual interviews are probably the most widely used method of exploring the respondents' perspectives. The interview may be relatively structured, with a long list of questions and prompts to encourage responses, or unstructured with perhaps only a brief list of topic areas. 'Narrative' interviews are similar to the methods used in oral history and can be a valuable approach to gain understanding of the events surrounding health behaviours and beliefs, decisions to consult, and experiences of diagnosis and treatment(10)(11). One of the benefits of the method - which requires audio or video tape recording - is that the resulting account is likely to reflect the person's priorities and concerns more faithfully than a structured interview. This is because the more structured the method of data collection, the more likely it is to reflect the agenda of clinicians or researchers, rather than the issues that are most pertinent to the participants. This is particularly important if the results are to be presented as representative of the lay perspective. A well conducted qualitative study will almost always identify some issues that were unanticipated at the outset. Flexibility in data collection and an analysis method that enables emerging issues to be explored, are seen as advantages of a qualitative method. Decisions about who is asked to participate, how they are asked, where they are asked and how the respondents perceive the nature of the enquiry inevitably affect the nature of the contribution. These are fundamental issues in qualitative research and are well discussed in the methods literature(12)(13).

Research interviews are so widely used in qualitative health research that one could be forgiven for concluding that they are the only available method. However, where it is possible to do so, observations of naturally occurring behaviour add an important perspective. Researchers who are intending to use interviews could benefit from considering whether observation may be more appropriate, or may add another dimension to their study. Britten's recently published work, exploring misunderstandings in the consultation, combined observations of consultations, followed by individual interviews(7).

In summary, when the aim of the research is to try to identify the beliefs, perspectives and priorities of the participants, a relatively unstructured, qualitative research method is usually most appropriate. The next section of this paper describes a new project that is using qualitative methods to explore people's beliefs about health and disease, and experiences of illness and treatments.

DIPEx a database of individual patients experiences

Some qualitative researchers are working within multi-disciplinary groups to develop ways of collecting, analysing and presenting research to help inform health professionals, researchers and policy makers about the patients' perspective. DIPEx (database of individual patients' experience) is a new project which is collecting and analysing narrative interviews to identify the issues that are important to the patients.

The central aim of the DIPEx project is to improve understanding of people's experience of illness(14). The resource is being developed as an internet website and CD Rom for patients, family carers and for teaching health professionals. Each site includes a collection of interviews, information about the condition, links to further resources and a site index. The information is prepared using the best current evidence about the condition and the available treatments. This includes describing the prevalence of the condition, information about causes and details of the effects of treatments, including complementary therapies. Questions and answers are presented, based on the issues that were identified in the interviews and in the literature. In the hypertension study, many of the questions concerned causes, prevention and the effects of lifestyle modifications such as weight loss, dietary salt reduction and increased physical activity. The 'further resources' section includes details of support groups, other useful websites and details of books and articles about the condition.

Each module of the site will include a wide range of material relevant to one condition (www.dipex.org). For example, the hypertension site (due to be completed in Spring 2001) contains video and audio taped interviews with a sample of forty men and women with high blood pressure, purposefully selected to represent a wide range of experiences. Participants were sought through GPs in different parts of England and include a range of social class and ethnic backgrounds, people whose high blood pressure was discovered in different ways and experience of all of the main medications used to treat hypertension. The interviews were all semi-structured and covered the story of each participant's hypertension, as they recalled it, from when it was first discovered. The interviews were analysed as a complete set, to identify both anticipated and emergent themes. These included: how the hypertension was discovered, images of the kind of person likely to have hypertension, symptoms, ideas about causes and family history, side-effects of medication, attitude to taking the medication, sources of information, fears about stroke and heart disease, use of complementary therapies, questions and concerns about high blood pressure and attempts to lose weight and change diet.

The methods that are being used to collect interviews for the DIPEx focus on the patient's story, structured according to their preference. Because the data collection is relatively unstructured and not shaped by a single research agenda, the interviews can be used as a resource for a wide variety of investigations, including issues that are important in developing health promotion strategies, such as the perceived benefits of lifestyle changes and concerns about taking medications. Figure 2 presents, as an example, perceptions about taking anti-hypertension medication. It is, of course, quite possible to feel that the medication protects, while also having considerable concerns about side-effects. However, some participants questioned whether they really needed to take the drugs and suspected that they might have 'white coat hypertension' that only manifested itself in the clinic. Participants with home blood pressure monitors had been able to reassure themselves that they really did have hypertension, which removed this concern about taking tablets unnecessarily. For some people, dislike of taking medication was serious enough to make them consider not take any drugs:

And I am seriously thinking of not taking the tablets .. I do feel … I wake up in the morning and I actually feel quite good and then I take the pill and I seem to lose energy. I really … I don't know if I am imagining it but I seem to not have that life force that I had before I took the pill and I don't know whether that's because I secretly don't like drugs - do you see what I mean? And obviously I'm not stupid and if I knew my risk was incredibly high I would just say well I'd rather be here than not be here, rather be here feeling a bit rough than not be here, and I would take them. RM14

The analysis of the themes which is presented on the DIPEx site is inclusive, taking account of all of the interviews in the collection. A short summary of each theme is presented, illustrated with video, audio and written clips from the interviews. Figure 3 shows an example of a summary of the theme dealing with attempts to reduce salt consumption, with a selection of illustrative quotes from the hypertension interviews. Other summaries will be available on the website, while the full content of the interviews (for which copyright approval has been granted) will be available to other researchers, on request.

Conclusion

Lay involvement through representation on advisory groups, and in shaping local and national policies, has many benefits that cannot be supplanted by the findings of qualitative research. There are now a number of examples of excellent practice of consumer involvement, as well as guidelines to help ensure that the participation is fruitful. There are concerns about the predominately middle class, or otherwise atypical, backgrounds of user representatives and Williamson has pointed out that patients will not usually be able to speak for other patients, 'partly because they will not know what their views are, and partly because their analyses will be confined to their own experiences'(9). We should, however, be cautious of double standards since the professional members of advisory groups are usually welcomed for their specialist, rather than typical, interest in the area.

Both lay and professional members of advisory groups could benefit from identifying relevant qualitative research that may provide access to the perspectives of a far wider range of people than are likely to participate as individual members of advisory groups. Qualitative researchers will help if they present their findings in ways that are useful and understandable to the general public, while maintaining rigorous methods of collection and analysis.

The development of health promotion strategies for primary care requires an understanding of a wide range of lay perspectives. Qualitative research with smokers or people who have tried to lose weight, or those who have experienced barriers to behaviour change, can help inform the development of appropriate approaches in primary care. These may include the identification of target groups to whom to offer an intervention (e.g. free nicotine replacement therapy for the partners of pregnant women) or training staff to recognise that their rapport with a patient may be damaged if they are indiscriminate or heavy handed when dispensing advice(5).

The DIPEx database has the potential to be valuable resource for those who are seeking to identify user perspectives. Because the methods include collecting purposeful samples of narrative interviews; analysing the data in a systematic manner; and presenting summaries of the main themes from the interviews, DIPEx will provide access to a broad range of patients' perspectives. Current and planned DIPEx projects can be viewed on the website (www.dipex.org) and include experiences of hypertension, prostate cancer, cervical screening, breast cancer, colorectal cancer, the carers of people with dementia, smoking cessation and rheumatoid arthritis.

Acknowledgements
Thanks to Professor Jurgen Pelikan and Dr Karl Krajic for inviting me to present this paper at the European Conference on Health Promotion in general Practice and Community Pharmacy: experiences and perspectives 10-11 November 2000, Brussels; to the participants in the DIPEx interviews; to Rachel Miller, who interviewed and analysed some of the data quoted here, and to the rest of the DIPEx group (Andrew Herxheimer, Ann McPherson, Sasha Shepperd, Janet Caldwell) for permission to use our data in this paper. The Department of Health funded the hypertension study. Sue Ziebland is employed as a senior research fellow for the Imperial Cancer Research Fund General Practice Research Group.


Appendix 1: Involving the consumer perspective

Individual involvement

- On advisory and steering groups
- Voluntary organisation committees
- Funding and research agencies
- Consumer groups, e.g. Cochrane Collaboration

Research on consumer perspectives

- Questionnaire surveys
- Semi-structured telephone interviews
- Focus-group discussions
- In-depth individual interviews
- Observations and interviews

Appendix 2:

Perceptions of taking anti-hypertension medication

Benefits
- Feeling protected against stroke
- Migraines stopped, symptoms reduced

Concerns
- Side-effects, now and in the future
- Dislike of taking any drugs
- Questioning the necessity of taking any drugs
- Not being able to forget about the hypertension

Appendix 3

Salt reduction

Amongst the various lifestyle changes that people described having made, or having tried to make, attempts to cut down on salt were prominent. Awareness that there is a link between salt intake and high blood pressure is widespread, and several people told us that doctors, nurses and friend(s) or family had advised them to cut down. Others had read similar advice in leaflets or books. However, not everyone felt that the advice was sufficient motivation to stop using salt. Some salt users had continued eating as much as ever, unwilling to sacrifice their enjoyment of food and other attempts to cut down had not succeeded. Those who had managed to stop using any added salt reported that their taste had changed and salty food was now quite unpalatable. Others had cut down in some ways - for example by reducing added salt at the table, or avoiding salt in cooking - but had clearly retained a taste for the seasoning and described certain foods as needing salt, such as mashed potatoes and tomatoes. Several said that they had not needed to change the amount of salt in their diet because they used very little or because they had given up using it many years ago or, in the case of one, because she had read that salt reduction should be avoided when using her anti-hypertension medication Trasidrex.

Even those who have successfully cut down on added salt in cooking or at the table could still be consuming a lot. Those who had tried to reduce salt reported some of the difficulties in identifying high salt foods and had become aware of the high salt content of certain bottled mineral waters and of the poor labelling of low salt foods compared with low fat foods in the supermarket. One drew attention to the high salt content of many convenience foods and slimming foods - which could cause problems to those who are trying to lose weight and cut down on salt after discovering their hypertension. While one had given up eating marmite on toast as part of his salt reduction plan, another who described himself as not eating salt, reported that he uses a lot of soy sauce on his food.

Some of those who had been successful in reducing their salt had made other changes at the same time - for example cutting down on fat intake or cutting down on alcohol and giving up caffeine. While it is possible that making several changes to the diet at once is easier than continuing to eat the same foods without the habitual seasoning, these individuals may have regarded their health as a particularly high priority.

Examples from interviews

I probably reduced our salt intake by about half by for example just not putting salt into vegetables when I cook them and we tend not to add salt to anything. But there's certain things I think that just aren't worth eating if they haven't got salt in so, but I was quite surprised at how much I could reduce it and I, it took me a while to get used to it but I did get used to it. I was told I would and eventually I did, it's true I did.

How long did it take you?

It ran into months rather than days or weeks and now I can eat them either way and am quite happy either way. SZ13

Has anybody ever suggested any changes to your diet, has anybody ever suggested that you should use less salt or anything? Do you use much salt?

We have salt in our food, but I just cannot see why a mineral like that makes a difference to your life. I mean people have always said ‘Oh use sea salt,’ but I don't, I used to put a lot of salt on my dinner but I was told once, I don't know ‘Do you use a lot of salt in things?’ but I can't say it's excessive, no.

So nobody has ever suggested that that might be … ?

Well if I cut salt out I think food is very bland without cooking with something. But what I would like to know is how some people suffer with high blood pressure and some people don't. It must be your make up, the make up of the person that's all I look at. Like people who have heart attacks and people who have cancer you know, there's got to be something in your make up that makes you have that particular ailment. SZ08

Well as soon as she said I've got high blood pressure, self-help I'm a great believer so I'll try and cut down on the alcohol and I'll try and cut salt out. I know salt's bad. I used to eat quite a lot of salt, really - probably far too much for my own good. We always had a salt-pot on the table. We had plenty of salt. To be honest, I don't miss it either. We still use it in cooking but not as much as what we used to do.

Did you find your taste changed after you stopped eating salt? Do you find meals sometimes too salty when you eat out?

I don't know. If you put salt on a tomato it's totally different. In some ways, I still miss it actually; some of the flavours are probably better. But it is getting used to. It's like when you give up sugar, isn't it. In tea, at first. Once you get used to it, you couldn't go back to it. Well, that's my opinion, anyway. SZ07

So again until I came to England, where we had lots of fast foods and for convenience I was having, working as a nurse and putting in all the hours and sometimes you'd have to take meals to work if there wasn't food provided for you and it was always the convenient things like the packet foods and whatever. And then luckily they were bringing in slimming foods. And you look at them and you think okay well perhaps you can have slimming foods but I noticed that I was having a lot of those and not really checking my sodium but in fact I was having quite high sodium content. Because whilst I think that the food was slimming foods they had so much salt in them …RM17