- S. Papagrigoriadis* and A. Koreli
Department of Colorectal Surgery, King’s College Hospital, Denmark Hill, London SE5 9RS, UK
*Correspondence to E-mail: firstname.lastname@example.org
Eubios Journal of Asian and International Bioethics 14 (2004), 135-138.
Interviewing is increasingly used in health services research to examine in depth the outcomes of treatment, quality of life issues and other health parameters. In this article we examine several problems generated when researchers use interviewing to explore life threatening issues such as cancer and other terminal conditions. We discuss whether the interview itself can be considered as an “intervention”, whether the interviewing process may have a detrimental effect on the patient, and situations where the interview can create distress or disrupt the patient’s coping strategy. The research interview may partially substitute counseling and the researchers may face ethical dilemmas when they are at the same time involved in the clinical care of the patients they are interviewing. The research interview is a complex issue that needs a careful and well designed approach to achieve the research objectives without harming the patient or the clinician-patient relationship.
Keywords: Interviews, qualitative research, life threatening issues, sensitive issues, doctor patient relationship, medical ethics, quality of life
Qualitative Research has its origins in Social Sciences but it has moved into the area of Health Sciences. Although it has mainly flourished within Nursing Research, there has been recently an increasing emphasis within Medical and Surgical research on Quality of Life (QoL) as an outcome measure of medical treatments and even surgery. Disease - specific QoL instruments are constantly being designed and tested in all medical areas. Patient interview is a key technique in developing those instruments. Clinicians are increasingly likely to be involved in clinical studies that contain QoL assessment, with structured or non structured interviews, alongside collection of clinical data. Clinical data are correlated with QoL data to produce the research findings. The same attention to ethical issues that is required for the design of the clinical part of the studies is also required for designing and implementing the interviews with patients. Interviews can be “straight forward” and risk free if the research object is benign disease. However, when the research object is a life – threatening disease, such as cancer, then more ethical issues arise. In this article we attempt to address some ethical questions and possible areas of risk when we interview patients with life threatening conditions for research purposes.
Interviewing the patient is the first step of any medical assessment. Taking the patient’s clinical history is in fact an interview. This interview aims to reveal the patient’s symptoms and problems and provide the physician with the necessary information in order to decide on diagnosis and treatment. During follow-up after any treatment, interviewing is part of the assessment of outcome: it aims to reveal if the patient is satisfied with the result of the treatment. Patients’ views on their care are a way of measuring the outcome of treatment. Quality of life assessment is important for life threatening conditions that require radical treatments. Those treatments often have an adverse effect on quality of life.
Apart from its role in clinical practice for assessment of treatment results and quality of life, interviewing is also a powerful research tool that allows insight into patients’ experience of disease and treatment.
Patients evaluate disease symptoms, short and long term problems, in a different way than physicians. A non-structured (without pre-designed questions) or semi-structured (based on pre-designed questions but allowing freedom to the patient to add other information) interview can help getting the patient’s perspective on the risk of adverse outcomes- death may be only one of them. Other adverse outcomes that may actually be more important to patients may have to do with inability to work, fear of social isolation, concerns about the fate of their children and others.
Each patient finds an individual “coping strategy” for those risks[i] .This strategy does not emerge instantly. Patients go through phases of reaction to the diagnosis. There is a denial/anger/acceptance model described by Gleser and Ihilevich[ii]. Lazarus and Folkman[iii] have described coping with risk as a constant behavioral and cognitive effort.
While interviewing on life threatening issues we must be aware of potentially different perspectives in order to look for it through open questions and probing. A research interview will often be used to assess the patient’s quality of life. This can be a difficult task. Often established treatments have side effects that are under-reported[iv] . On the other hand, patients with apparently bad quality of life, i.e. paraplegics, patients with colostomy or chronic diseases, have adapted to their lifestyle, draw some satisfaction fro their life and are not prepared to trade off quantity for better quality[v] .
One of the serious ethical considerations when designing a study with interviews on life threatening issues is whether the research process might affect the management. The patient may reveal personal or psychological problems that have not been recognized or appeared since last clinical review. The researcher will be ethically obliged to inform the treating physician, thus having an impact on management[vi] . This phenomenon will be more significant if the researcher is medically trained and thus able to detect easily clinically important information. While no one can argue with that from the ethical point of view, if it happens too often it might bias the findings, particularly if the design of the study involves repeat interviews with relatively long intervals in between.
One might claim that if they have accepted to participate patients feel confident that the interview process will not cause them any distress. However patients may feel they have to participate out of obligation to treating physicians while they are not entirely satisfied with the process. Others may participate because they feel they might obtain disclosure of additional information. A number of patients and relatives (10-15%) ask the researcher straight forward clinical questions[vii] . If the researcher is medically qualified they have to choose between refusing to answer and the risk of disclosing information that might be confusing or upsetting. A refusal will undermine the patient’s trust in the researcher and might cause people who are not familiar with research methodology to suspect that information is deliberately withheld from them. Obviously this will have a negative effect on the amount and quality of information disclosed during a non structured interview. On the other hand if the medically qualified researcher attempts to discuss the clinical details of disease and treatment there is risk of providing information that does not coincide entirely with that given by the treating physician. The result will be confusion and apprehension of the patient and difficulties with the management as well as with the clinicians-researchers relationships.
An “in depth” interview may disrupt the patients’ coping strategy. This is particularly true if the patient is still in the phase of denial or if they are not fully aware of the prognosis or other details of their condition. Although all patients participating in a study are supposed to be fully aware of their diagnosis and the relevant details, research has shown that quite a few of them are often unaware of how bad their prognosis isiv. There are a number of reasons for that: They may have been told explicitly but medical terminology as well as the details of diagnosis and treatment are often too complicated, particularly for patients of low education. Rejection and denial causes some patients to selectively forget the “worse” news such as the option of chemotherapy after surgery, the possibility of a colostomy or complications etc[viii] . Typically, while some patients with cancer want to have an estimate of how long they are expected to live, others avoid carefully to ask that question even at very late stages of the disease. Often their physicians may be aware of this but avoid to “force-feed” the details for humanitarian reasons. In certain Asian, Arabic or Mediterranean cultures the family reinforces this phenomenon by being over-protective. A way to find out indirectly how much they know is to start with questions about recollections from the time of diagnosis and treatment planning.
Closed questions are fast in the field and can be coded without effort afterwards. They trace easily the main points of interest of the researcher and are amenable to statistical analysis. Closed questions can be carefully phrased in advance to avoid causing stress, they can also be tested in pilot studies to ensure compliance with cultural factors of the patient population. Therefore interviews with closed questions represent the “safest option” from the ethical point of view. However they provide only a limited degree of options and thus may miss important aspects of the patient’s problems if those have not occurred to the researchers or have not emerged during the pilot phase of the questionnaire. Open-ended questions are time consuming both in the field and in the office since they require a major coding effort. However they give the best chance of revealing the depth and variety of all the problems that might otherwise escape. Semi- structured interviews with both open-ended and close-ended questions help to reveal much more then questionnaires[ix] [x]. Inevitably, though, they carry more risk for unexpected directions of the patients’ answers and the researcher must have good personal communication skills and experience in order to handle unexpected reactions with minimal stress for the patient.
Sometimes there rises the question of relative presence: some patients are poorly and feel safer with a relative beside them to monitor the interview in a protective way. Relatives are variable. They can stay in the background or they can take over the interview and handling them tactfully requires skills and experience of the researcher. Family relationships play a role and the researcher must have the skills to distinguish between a supportive relative and an overprotective dominant member of the family who may influence the patients’ answers and alter the findings of the research project.
The majority of patients (85-95%) are happy to participate and remain happy afterwards[xi] [xii]. Some use it as an opportunity to express gratitude towards the physicians while on the other hand others use it as an opportunity to complain about unsatisfactory management. A number of patients describe the interview experience as an opportunity of self-reflection and insight. However a small group (~5%) can find the interview very disturbing and regret itvii. It has not been possible to identify in advance who is at risk for this emotional reaction so the researcher has to be alert to this possibility for early recognition during the interview process.
Although not designed for this purpose the interview can be used as counseling by patients and families. If successful this can be beneficial. However counseling requires appropriate training and personality skills. Furthermore, counseling is not automatically acceptable and beneficial by all cancer patients and experience and discretion are important[xiii] .
Physicians often tend to ask close-ended, directed questions[xiv] . The reason for this is that they try to obtain a “working diagnosis” and because of time pressures. Studies with doctors and nurses found that both use “distancing tactics” when talking to terminal patients in order to maintain their emotional survival[xv] [xvi].
Interviewing as part of clinical management is subject to litigation risks. Litigation against doctors is increasing and is now starting against nurses as well. Informed consent documents are becoming increasingly extensive and technical. Since 1998 the GMC has allowed patients to tape record the clinical interview and the consent process. An increasing number of the patient complaints received yearly by NHS trusts are about verbal interactions, misinterpretation of information or non- verbal communication between staff and patients.
Doctors and nurses may consciously or unconsciously adopt defensive and distancing interviewing attitudes to ensure protection from litigation at the expense of empathy and encouragement.
Interviewing on life threatening issues contains ethical dilemmas and risks of psychological harm to the patient. It also carries a risk of a negative effect on the doctor-patient relationship. The ethical issues we examined in this article should be taken into account when designing structured or non structured interviews and questionnaires for research purposes. The researchers in the field should have understanding of those issues and appropriate training and experience as well as inter-personal skills.
[i] Watson B & Heyman B (1998): Risk and coping with diabetes, p: 187-198, in: Risk health and healthcare, Ed. B Heyman, Arnold.
[ii] Gleser GC and Ihilevich D (1969): An objective instrument for measuring defence mechanisms. Journal of Consulting and clinical psychology, 33, 51-60.
[iii] Lazarus RS and Folkman S (1984): Coping and adaptation. In Gentry, WD ed. the handbook of behavioural medicine. Guilford.
[iv] Davies E, Clarke C, Hopkins A (1996): malignant cerebral glioma-II: Perspectives of patients and relatives on the value of radiotherapy. BMJ 1996 Dec 14;313(7071):1512-6.
[v] Stiggelbout AM. de Haes JC, Kiebert GM, Kievit J., Leer JWH (1996): Tradeoffs between quality and quantity of life:development of the QQ questionnaire for cancer patient attitudes. Med Decis Making;16:184-192.
[vi] Plant H (1996): Research interviewing. Palliat Med 1996 Oct;10(4):339-41.
[vii] Davies EA, Hall SM, Clarke CRA, Bannon MP, Hopkins AP (1998): Do research interviews cause distress or interfere in management? Journal of the Royal College of Physicians of London Vol. 32;5:406-11.
[viii] Miles BJ, Giesler B, Kattan MW (1999): Recall and attitudes in patients with prostate cancer. Urology 53(1):169-74.
[ix] Oppenheim AN (1998): Questionnaire design, interviewing and attitude measurement. Pinter, London.
[x] Bowling A (1997): Research methods in health. Open University Press, London.
[xi] Funch DP, Marshall JR (1981) Patient attitudes following participation in a health outcome survey. Am J Public Health 71(12):1396-8.
[xii] Fallowfield LJ, Baum M, Maguire GP (1987): Do psychological studies upset patients? J R Soc Med 80(1):59.
[xiii] Moynihan C, Horwich A, Bliss J (1999) Counseling is not appropriate for all patients with cancer. BMJ 318(7176):128 .
[xiv] Beckman HB, Frankel RM (1984): The effect of physician behavior on the collection of data. Ann Intern Med 1984 Nov;101(5):692-6.
[xv] Maguire P, Faulkner A, Booth K, Elliot C, Hillier V (1995): Helping cancer patients disclose their concerns. European Journal of Cancer Vol.32A;1:78-81.
[xvi] Espinosa E, Gonzales Baron M, Zamora P, Ordonez A, Arranz P (1996): Doctors also suffer when giving bad news to cancer patients. Support Care Cancer 1996 4(1):61-3..
[xvii] Epstein RM, Morse DS, Frankel RM, Frarey L, Anderson K, Beckman HB (1998): Awkward moments in patient-physician communication about HIV risk. Ann Intern Med 1998 Mar 15;128(6):435-42.
[xviii] Maguire P (1985) : Barriers to psychological care of the dying. Br Med J (Clin Res Ed) 1985 Dec 14;29(6510)1711-13.