Over the past decade there have been several published articles acknowledging as good practice the involvement of patients in providing an evaluation of health services (Brooker, 1997; Chambers, 2000; NHS Scotland, 2000). The NHS Plan(Department of Health, 2000) recognises that old-fashioned demarcations and barriers between services still exist. There is a need to create a patient-centred NHS with different parts connecting efficiently.
This paper reports on a qualitative observational study that examined the experiences of elderly patients and their families on admission to a hospital elderly care unit. The study design is described in detail and some issues for qualitative researchers in the hospital setting are discussed. This description may be useful for others wanting to conduct similar research in the hospital setting.
The research question was: ‘What is the experience of patients and their carers during the admission process to a care of the elderly rehabilitation/ medical assessment ward?’ Various methods of answering the study question were investigated and, initially, some of the more traditional methods were rejected as inappropriate, including a randomised controlled trial, questionnaires, interviews and diaries.
The method decided upon was the use of observational research to produce descriptive material that would allow the researchers to understand the actual process and meaning of admission to older patients. As described by Carter (1996), a descriptive design is useful when the issue being studied is not amenable to statistical analysis. The researcher is concerned with observing, describing and documenting occurrences which take place in everyday practice and can inform future practice.
Observation provides access to actual rather than reported behaviour, especially where patients are either reluctant or unable to describe their actions (Lynes, 1999). This method, however, is not without its problems, as data collected can reflect the observers’ preconceptions and prejudices (Mays and Pope, 2000). Qualitative observational methods have particular strengths in terms of uncovering social processes and the nature of personal experience (Murphy et al, 1998). In non-participant observation, the observer maintains the role of onlooker and so does not become a member of the group. However, it is difficult to maintain a completely passive role (Lynes, 1999).
As nurses, the two researchers were familiar with the ward environment and could participate in ‘short-life work’ such as making beds, assisting with meals and fetching and carrying for others. It was useful to have a physical task to occupy the time when the patient being observed was behind screens or asleep. Participation in ward activities also helped to make the observers’ presence less obtrusive.
In a busy ward there can be an ethical dilemma between the wish to assist busy colleagues and the responsibility of collecting true observational data. This is not an uncommon problem in qualitative observation (Gans, 1982).
Figure 1 shows the research method.
Several factors had to be negotiated before beginning the study.
FundingFunding was required to allow the nurses involved protected time for the research. Protected time is essential to carry out this type of study, as it is impossible to conduct in-depth observations during a clinical working day. One-year funding was obtained from the local primary care development fund.
Literature reviewA literature review was commenced before the study began and continued throughout. The following databases were searched: Cinahl, Medline, ASSIA, SOSIG, CHAIN, Healthgate, Cochrane, and the National Research Register. Hand-searching of journals during frequent visits to the nursing, medical and sociological sections of the university libraries uncovered information not yet on the databases. Information was also gained by networking through the National Research Initiative for Scotland and the Royal College of Nursing Research and Development Co-ordinating Centre.
Ethical approvalPermission from the ethics committee was required before the study could start. Permission to observe the admission process was given only once one of the researchers had attended an ethics committee meeting to discuss and clarify how this could be done without adding to the stress of the situation (Byrne, 2001).
The two researchers visited various staff groups to explain the aim of the study and elicit staff views. While the study was based on the views of patients and carers, these cannot be divorced from the views of the staff (Chambers, 2000). This interactive process of research definition and clarification was helpful in terms of access, goodwill and analytical themes relating to the data.
Observation is time consuming, so given the constraints of the study, timetable and budget the study sample was realistically small, consisting initially of two patients from each of the six wards within the unit. Four patients were later added to this sample to clarify experience over the hospital/ community interface. The study did not rely solely on observation. The integrity of the observations and initial conclusions drawn from them was checked with the patients and carers through very short, semi-structured interviews on the evening of admission and longer ones once the observation had finished (24-36 hours after arriving in the ward). Through these interviews, the researchers aimed to clarify reactions noted during observation and gain a greater understanding of the emerging themes. The admitting staff were also interviewed and asked for their view of the admission - had it gone smoothly, was all relevant information readily available, had there been frequent interruptions?
The field notes of the observations were transcribed within 24 hours of the observation ending. Early in the research process the two researchers met with the supervisor, who was an expert in qualitative research. Together the emerging themes and concepts were clarified and topics for further literature searches noted. Every observation produced a large quantity of data and this, together with the ongoing literature searches and reading, at times seemed overwhelming. Towards the end of the time allocated for observations there remained several concepts still requiring clarification. As time did not allow for further observations the researchers devised semi-structured questionnaires that were distributed to patients and their carers. The concepts targeted in these questionnaires were those of information (given and required), the admission process (timely and welcoming), and identity maintenance (name used and involvement in planning care).
The researchers’ preconceptions were that the admission process was tiring for patients and carers, with many different professionals demanding often duplicate information. The outcomes showed that some patients, however, found the attention given during the admission process stimulating rather than tiring. The more strongly emerging concerns were with information and communication, and with maintaining the identity and relationships of both patients and carers while acknowledging their expertise. From these concerns a theory of what constitutes a ‘good’ admission was built and tested by the production of 14 good-practice recommendations based on the study findings and supported by the literature review. (Examples of the recommendations are given in Box 1). These recommendations were presented to a group of 20 stakeholders who were chosen to represent the team of people involved in the admission process and consisted of doctors and nurses from primary and secondary care and staff from the social work department, ambulance service and professions allied to medicine, together with service managers, patients and carers.
Each recommendation was discussed and reworded if necessary until consensus was reached on its acceptance. Also included was a discussion of possible implementation strategies for these recommendations.
Issues for the qualitative nurse researcher
ValidityQualitative research should be assessed according to its validity, defined as the extent to which the account accurately represents the social phenomena to which it refers (Murphy et al, 1998). One method of enhancing this validity is triangulation - looking at the study situation from different angles and comparing the views in order to obtain a full and complete picture of the area under study. Hammersley and Atkinson (1983) have suggested three methods of triangulation. All three were included in the study:
- Data source triangulation. This was addressed by collecting data from staff, patients and carers. All types of admissions to the unit were observed, comprising same-day admissions from GPs, planned admissions, respite and shared care patients and transfers from the acute unit
- Triangulation between researchers. A district nursing sister and a ward charge nurse collected the data under the guidance of an experienced qualitative researcher and social scientist
- Technique triangulation. The observations were augmented and clarified by interviews with patients, carers and staff.
The complex nature of the individuals studied in qualitative research can reveal contradictions in the data collected and thus weaken the study results (Docherty, 2000). Docherty also states that, when using a combination of different research methods, usually including qualitative and quantitative approaches, all should have equal weighting. Some methods used in this study were perhaps midway between these two approaches.
GeneralisabilityIt was not the aim of the study to generalise the findings to other units, but to use the findings as a basis for the production of local guidelines. However, it was important that our sample included a full range of possible cases so that conceptual generalisations could be made (Mays and Pope, 2000). The original sample was extended by adding four more patients to the sample in order to clarify and challenge the themes emerging through analysis. The aim therefore was not generalisability but the ‘relatability’ of the findings to the research unit and the transferability of the method or the findings to other similar units (Bell, 1993). Relatability is the extent to which someone working in a similar situation can relate their practice and decision-making to the research findings. Such an aim is not unusual in this type of research because ‘qualitative research frequently does not seek to generalise to a wider population for predictive purposes, but seeks to understand specific behaviour in a naturally occurring context’ (Pope and Mays, 2000).
ReliabilityReliability refers to the degree of consistency with which instances are assigned to the same category by different observers or by the same observer on separate occasions (Hammersley, 1992).
In qualitative research, reliability cannot be achieved in the traditional scientific sense where an unchanging reality and stability can be assumed. Research in a natural setting makes replication difficult. People’s behaviours change and the health service is in a period of constant change at present. It has been argued in the past that the traditional emphasis on reliability is not applicable to qualitative research (Guba and Lincoln, 1989; Kirk and Miller, 1986). If a qualitative research study is to have a practical application, however, those who are going to put the findings into practice must have confidence in the research findings and reliability is therefore important.
‘Obviousness’It can be argued that the results of qualitative research are mere truisms or common sense, which most people would know without having to go to the trouble of doing the research (Gage, 1993). What is common sense to one person can be a mystery to another, however, and this study found that, while a recommendation might already be in place in one area or partly practised by individuals in another, overall practice could be patchy. It cannot be assumed that all practitioners work in the same way and this has implications for equity in care. As practising nurses, the researchers were concerned to find ways of implementing the recommendations in order to eliminate inequalities in care received and to incorporate patient and carer perspectives into the care giving. These variations were acknowledged by all present on the consensus day and the importance of all 14 recommendations was agreed.
Translating research findings to practice
Nurses should be concerned with giving the best care possible to each unique, individual patient (RCN, 1993).
The professional knowledge base of nursing includes scientific theory learned from lectures, books and journals; experiential knowledge gained through practising as a clinician; and personal knowledge about an individual patient or situation. Professional judgement is the combination of all three types of knowledge used to determine what would be the best course of action in a given situation (Rolfe, 1998) The good practice recommendations from this study are a useful aid to the application of professional judgement during the admission process; they increase the confidence with which the existing knowledge base can regarded and applied.
In order for research to be translated into practice, nurses must be aware of the research, and accept the responsibility that, where research is applicable to care, it should be put into practice (UKCC, 1992). In this localised study staff opinions were sought before the research began. Staff were also included in the group of stakeholders who finally formed the consensus group for the acceptance and refining of the good practice recommendations.
Change is easier to implement when research has a shared governance approach in which staff are involved and have power to make decisions (Doherty and Hope, 2000). A restructuring of documentation in which all staff felt they had ownership prompted this study. Guidelines for the admission process were not available and the review of patients’ and carers’ experiences was seen as the first essential step in their production. Although staff were informed about the research and had a positive attitude towards it this does not mean that the findings will be automatically absorbed into practice.
Some of the good-practice recommendations have been relatively easy to implement or have fitted naturally into the ongoing development of the service. Other recommendations require a change of culture. The inclusion of carers in care giving is practised to a limited extent in the preparation for discharge. The extension of carer participation in care to being an integral part of the whole process of rehabilitation rather than as a preparation for discharge will need a cultural change.
Funding is being sought for an implemen-tation study that will include facilitating this culture change. Over the past decade the culture and emphasis of rehabilitation practice has been changing from a medical model, focusing largely on disability and physical functioning (Nolan and Nolan, 1997), to one in which older people are recognised as experts on themselves. The basis for changing culture is therefore already underway and can be used as a foundation for further change. As Reed et al (2000) states: ‘People have more confidence moving to the future when they carry forward parts of the past.’
Using a variety of techniques (observation, interviews, brainstorming and questionnaires) this study was able to describe patient and carer experiences during admission to hospital. Common concerns were identified and these have informed the production of 14 good-practice recommendations that will be used for standardising and improving care.
It has been recognised that the initial one-year funding will not provide the support needed for implementation of the recommendations, which require a change in culture. Funding is currently being sought to extend the study to a comparative intervention study over three sites in Scotland.
There have been several practice developments within the unit since the study. Length of stay for patients is now predicted at the first ward round after admission and the results of these predictions are noted on discharge and audited. A patient diary has been introduced as part of the documentation and a ‘Getting to know you’ document is being piloted in one of the wards. Patients’ and carers’ expectations of the admission are now noted and their input is being documented in a more structured way in the care plans as they are revised.
- The authors would like to acknowledge the support of Tayside Primary Care Trust during this study and Tayside Primary Care Development Fund for funding the study.
Bell, J. (1993) Doing Your Research Project (2nd edn). Buckingham: Open University Press.
Brooker, D.J.R. (1997) Issues in user feedback on health services for elderly people. British Journal of Nursing 6: 159-162.
Byrne, G. (2001) Ethical approval. Professional Nurse 16: 1052-1054.
Carter, D.E. (1996) Descriptive research. In: Cormack, D.F.S. (ed.). The Research Process in Nursing (3rd edn). Oxford: Blackwell Science.
Chambers, R. (2000) Involving Patients and the Public: How to do it better. Abingdon, Oxon: Radcliffe Medical.
Department of Health. (2000) The NHS Plan: A plan for investment, a plan for reform. London: The Stationery Office.
Docherty, B. (2000) Using triangulation in health-care research. Professional Nurse 16: 926-927.
Doherty, C., Hope, W. (2000) Shared governance: nurses making a difference. Journal of Nursing Management 8: 77-81.
Gage, N.L. (1993) The obviousness of social and educational research results. In: Hammersley, M. (ed.). Social Research: Philosophy, politics and practice. London: Sage.
Gans, H. (1982) The participant observer as a human being: observations on the personal aspects of fieldwork. In: Burgess, R.G. (ed.). Field Research: A source book and field manual. London: George Allen & Unwin.
Guba, E.G., Lincoln, Y.S. (1989) Fourth Generation Evaluation. Newbury Park, Ca: Sage.
Hammersley, M. (1992) What’s Wrong with Ethnography? Methodological explorations. London: Routledge.
Hammersley, M., Atkinson, P. (1983) Ethnography: Principles in practice. London: Routledge.
Kirk, J., Miller, M. (1986) Reliability and Validity in Qualitative Research. London: Sage.
Lynes, D. (1999) Using observation for data collection. Professional Nurse 14: 315-317.
Mays, N., Pope, C. (2000) Qualitative research in health care: assessing the quality in qualitative research. British Medical Journal 320: 50-52.
Murphy, E., Dingwall, R., Greatbatch, D. et al. (1998) Qualitative Research Methods in Health Technology Assessment: A review of the literature. Nottingham: National Co-ordinating Centre for Health Technology Assessments.
NHS Scotland. (2000) Our National Health: A plan for action, a plan for change. Edinburgh: Scottish Executive.
Nolan, M., Nolan, J. (1997) Rehabilitation: realising the potential nursing contribution. British Journal of Nursing 6: 1176-1179.
Pope, C., Mays, N. (2000) Qualitative Research in Health Care. London: BMJ Publishing.
Reed, J., Pearson, P., Swinburne, S. et al. (2000) Going home from hospital: a collaborative study. In: Dickinson, A., Bartlett, H., Wade, S. (eds). Old Age in a New Age (Proceedings of the British Society of Gerontology 29th Annual Conference). Oxford: Oxford Brookes University.
Rolfe, G. (1998) Expanding Nursing Knowledge: Understanding and researching your own practice. Oxford: Butterworth-Heinemann.
Royal College of Nursing. (1993) The Value and Skills of Nurses Working with Older People. London: RCN.
UKCC. (1992) The Code of Professional Conduct. London: UKCC.