Ethical review of physiotherapy research.
In this article, we discuss a number of ethical issues for physiotherapists conducting research. Compliance with ethical standards expected of all health care professionals involved in research--including the requirement to submit research for independent ethical scrutiny--can usefully be supplemented by examination of the ethical reasons for these standards, and also reflection upon the reality of conducting physiotherapy research, often in the midst of clinical practice. This article seeks to explore these issues.
Discussion of the ethics of research focussed explicitly on physiotherapists is timely. Currently in physiotherapy, as in other professional health activities, there are demands for more research on the one hand, and for greater ethical awareness and compliance on the other. In a recent article, Henley and Frank (2006) recognise the drive for more and better evidence of the effectiveness of clinical practice, and the current pressure for compliance with ethical standards being led by publishers of journals. They recommended that 'Physical therapy journals should standardize ethical protections and make documentation of compliance a prerequisite of publication' (Henley and Frank 2006: p.499). The insistence that research should have been conducted in an ethical manner before it can be published gives researchers a strong reason for ensuring that their research does comply with standard ethical protections for as Henley and Frank describe, it is 'a powerful mechanism for encouraging ethical research' (p.504). They suggest that published research comply with three such protections, these are: seeking research ethics committee approval prior to carrying out research; getting informed consent from participants, and keeping information gathered about participants confidential (Henley and Frank 2006).
Physiotherapists not explicitly involved in research may consider a discussion about research ethics is irrelevant to them. However many physiotherapy treatments have evolved greatly from their original form and/or have been applied to new patient (and perhaps more vulnerable) groups in which this treatment has not previously been used. Questions must be asked about whether such treatment should proceed without supporting research that is able to assess the appropriateness of such treatment. This article is designed for physiotherapists engaged in research and those involved in the creation of innovative therapies.
Research ethics committee approval: The importance of independent ethical review
We know from history that research participants have not always fared well and concern for their welfare has, at times, been lacking (Beecher 1966, Paul 1988). There has been recognition worldwide in the light of this history that safeguards are required to ensure that the welfare of research participants is protected (Evans and Evans 1996). Central among these safeguards is that the proposed research should be reviewed by an independent ethics committee (World Medical Association 2000, Nuremberg Code 1946-49). The establishment of independent ethics committees in New Zealand (NZ) was in direct response to events at National Women's Hospital where research into the natural course of carcinoma in situ of the cervix was carried out from 1966 until 1987 (Cartwright 1988). The research was brought to the attention of the public by two journalists (Coney and Bunkle 1986), leading to a government-appointed inquiry chaired by Judge Silvia Cartwright. The Inquiry found that a research trial had been undertaken in which many women did not know that they were participants in research and that some women had died due to a failure to supply standard treatment (Cartwright 1988).
The primary researcher had taken his research to a hospital ethics committee (Paul 1988), but this committee had failed to recognise the potential harms and other ethical issues the research raised. Part of the explanation for this might be that it was not sufficiently independent or empowered to question and reject research. The committee was made up of many of the researcher's peers, and was supportive of the concept of professional freedom for clinicians (Paul 1988). As such, it lacked the objectivity required to identify the flaws in Green's research and also the perspective required to protect the study participants (Paul 1988).
The current system of ethical review in New Zealand is to a major degree the child of this experience (Nie & Anderson 2003). Where health-related research on patients is concerned, there are currently 6 regional committees (3 on the North Island and 3 on the South Island) and a multi-region committee (which reviews research going on simultaneously in more than one region) charged with the task of ethical review of research. These committees have a membership appointed by the Ministry of Health. Some human research--that which does not involve patients for example--may be reviewed by committees established within tertiary educational institutions (universities and polytechnics). Their membership is a mixture of community and institutional appointments.
Ethics committees are guided by a number of recognised and widely accepted ethical standards (for an account of these, see next section) and by the system of accreditation by the Health Research Council (HRC) Ethics Committee. The HRC Ethics Committee accredits ethical review committees on the basis of the Operational Standard (Ministry of Health, 2006). To achieve accreditation, committees must, for example, have the appropriate mix of community (e.g. non-health professional, or non-institutional) and expert representation to make effective ethical judgements. They also must have sufficient Maori representation, and an acceptable gender balance. Their processes and principles must be consistent with thorough and efficient review; e.g. they need to guarantee that each protocol receives sufficient consideration, and that mechanisms are in place to inform researchers of the committee's decisions. All of the Ministry of Health (regional and multi-centre) committees are currently accredited, as are a number of the major institutional committees.
In New Zealand, the ethical review system has two other elements: firstly Maori consultation is regarded as an important element of much research (Sporle and Koea 2004). This consultation is in addition to the review carried out by the committees, including their Maori members. Secondly, as far as health research goes, there is also what is called a locality assessment--which is, at least, a declaration signed by the relevant local health institutions that the research is practical in the locality and site planned (e.g. that the researcher has time to do the research, or that the local hospital has the facilities to enable to research to take place).
However, important though it is to get independent ethical review, approval of a research proposal by a research ethics committee should not be viewed by physiotherapists as the end of their ethical obligations nor a substitute for their own ethical awareness. Attention to ethical concerns should be a primary concern of all researchers throughout the time period of the research, and researchers must be alert to these.
Values informing ethical review
As was mentioned in the previous section, there are some widely accepted ethical principles which are understood to apply to research. In addition, there are a number of guidelines which express these values, which have been promulgated over the years by different organisations and committees: some of these will be mentioned briefly below after the consideration of the values themselves.
A well accepted framework of such values is that of Beauchamp and Childress (2001). Among the values they list which are most obviously applicable to research are: respect for autonomy; beneficence; and non-maleficence.
Autonomy refers to the idea that each individual should be in control of their own destiny and their own person (Beauchamp and Childress 2001, 57). Respecting this value, in the general context, means protecting people from those things, such as coercion and manipulation, which might threaten to undermine this control, and ensuring people are empowered to exercise this control, for example by making sure they have relevant information. In the research context in New Zealand, autonomy is realised most obviously in the right to informed consent--the right to decide upon the basis of good information whether or not to participate in research (Health & Disability Commissioner 1996). It may also be realised in respect of confidentiality of health information, since control over oneself may be perceived to extend to control over certain information about oneself (Ministry of Health 1994).
Beneficence refers to the aim to make things better for people in general. In the health research context this relates to the fundamental moral justification for health research, which is to ensure the development of effective and safe treatments for human ills, including of course physical treatments.
Non-maleficence refers to the avoidance of doing harm, and is often seen as rooted in the Hippocratic tradition (Campbell et al 2005). In the health research context, it is often argued that there needs to be some balancing of beneficence and non-maleficence: for the term research implies trying something out that has not been tried out previously, which implies that the potential to do good or harm is not yet understood. So, the people on whom the treatment is being tried out are, to some extent, 'guinea pigs'--albeit in most cases informed and consenting guinea pigs--who take on the risk of harm on behalf of and for the benefit of a wider group.
Amongst the guidelines which reflect these values, and apply to research and researchers in a general way, are those of the Health Research Council of New Zealand (e.g. Guidelines for Health Research With Children, HRC 2007; Guidelines for Researchers on Health Research Involving Maori, HRC 2008) and of the National Ethics Advisory Committee (NEAC) which has developed guidelines on conducting observational studies, audits, and related activities (NEAC, Dec. 2006) and is developing a parallel set of guidelines for intervention studies and innovative practice (NEAC, Work Programme). There are also international guidelines which apply in New Zealand, for example the World Medical Association Declaration of Helsinki (WMA, 2000).
The values of autonomy, beneficence and non-maleficence, and their general applicability to health research (including physiotherapy research), are well established. But in any actual circumstance in which physiotherapy research is conducted, all kinds of more or less hidden ethical dangers lurk. These may undermine the application of the accepted bioethical values. The aim of the remaining sections of this paper is to bring some of these hidden dangers to light.
Autonomy: some hidden dangers to informed consent
As stated earlier, obtaining the consent of a person to participate in research is plainly important. It may not be a requirement in all cases, as we discuss below. Consent, where it is required, as in the vast majority of cases, functions as a protection for participants, ensuring that they are not subject to any experimental or innovative treatment or other intervention which they do not want. Typically, in fulfilment of consent, potential participants must be adequately informed about what is proposed. This means writing an information sheet for participants which is clear, fully states the purpose of the research, any attendant risks, costs and compensations, as well as what the data to be collected are to be used for. Other information may include how data are to be stored, and for how long. Furthermore, participants are generally assured that they can withdraw their consent at any point in the research, without the need to give any reason, and without any penalty or come back.
What if consent can't be obtained?
Physiotherapists will from time to time want to carry out research on people who are not competent to consent. Where there is a choice, research should always be carried out in groups that are competent to consent. But this might restrict the ability of physiotherapists to develop properly evidenced therapies for some groups of clients, for example for those with severe brain injury associated with increased muscle tone, where the clients' inability to consent and special physical needs may be related. One possible way forward in a case like this is to ask would this person, when competent, have been willing to take part in research. A client's family may be a good source of information for answering this question. A highly detailed application to an ethics committee would be necessary and advice should be sought on proper practice.
While it is the case that there are some people who are incompetent to consent, care needs to be taken not to assume this. For example disabled people may be thought to be unable to give consent when in fact many may be perfectly competent. It may just mean that they need more time and more effort put in to explaining the research rather than just assuming that they lack the competence to consent. A rule of thumb in this area is that there should be a presumption of competence in all cases. Incompetence must be proved, never assumed.
Consent and relationships between physiotherapists and participants
A hidden danger in seeking someone's consent to take part in research lies in the surrounding context of physiotherapy practice and the relationship the physiotherapist has with participants. The context and the relationship have the potential to threaten autonomy even when other aspects of obtaining consent are in place (Tolich and Baldwin 2005). Many practicing physiotherapists, like other practicing health professionals, appear to have a readymade 'pool' of potential participants in their own clients. This is not to say that recruitment is limited to this pool. But in many cases the physiotherapist's own client base will be easy to contact, easier to introduce the subject of possible research participation to, and may well be easier to recruit since they may place a good deal of trust in the physiotherapist, and believe that the research must be worthwhile being part of just because he or she is conducting it.
But select ion of participants from a physiotherapists own client base should sound ethical alarm bells. One question might be how easy individuals in this client pool would find it to say 'no'? How far is their capacity to formulate and express doubts about the research and/or their participation in it undermined by their existing relationship with the physiotherapist? For example, a physiotherapist helping with rehabilitation during long-term care after an accident or illness may enter a long-term and intense relationship with a patient, through frequent, lengthy sessions. Moreover, the role of the physiotherapist in trying to return the patient to optimal function may mean that he/she has a particularly important actual and symbolic role in the patient's life. These factors, in a situation where the physiotherapist makes a request of the patient, may undermine the patient's ability to make a considered and independent decision.
An ethically problematic context such as this is not, however, necessarily a red-light to doing research. One possible way around this problem is for the physiotherapist to ask someone else to take on the role of obtaining consent to participate where the potential participants are clients. Where this is not possible, the physiotherapist might advise clients to think very carefully about whether they really want to participate, suggest they consult with family members before deciding, and make certain that they are aware that participation is by no means expected of them, and that if they say 'no' their care will not in any way be affected. Just allowing time for the decision ('I'll see how you feel about it again next week') may help the client reflect upon things in a truly independent way.
There are some populations whose ability to refuse consent to take part in research (or judgement) may be undermined by wider social attitudes and their reliance on others. This group may include prisoners, elderly people in long stay care and children. For example an elderly person in long term care may feel unable to refuse to participate if he/she believes the care will be affected by a refusal. Particular care needs to be taken with these groups that information is given appropriately and full opportunity given to refuse consent without any detrimental effect on care provision or treatment.
Confidentiality: promise and practice
Confidentiality is a protection for research participants in that it restricts access to knowledge gained about them as part of or simply during the research. The participant is usually assured that only those actually collecting data will get to know who the data came from, that the data will be kept hidden away from others, and that there won't be any way for a third party (such as a statistician or the reader of a journal) to be able to connect the data with an individual participant. These guarantees of confidentiality usually extend for the whole period of research and beyond. This can be understood to include the period of data collection and analysis, presentation and final storage (Ethics committees may demand a storage period of raw data of a number of years.) Where these safeguards of confidentiality cannot be offered, the participants should at least be told, so they can decide whether or not they are happy with whatever level of confidentiality can be offered.
Confidentiality is generally regarded as an important right for research participants, but it may be argued that some physiotherapy research is unlikely to threaten it. For example, quantitative physiotherapy research often involves objective measurements of variables, and efforts to generalise from the sample in the research to its wider population, according to statistical norms. This might include, for example, measuring the range of motion of a particular joint. This data would most often be useless for identification purposes on its own: if someone who was not intended to see it did, they'd be none the wiser about the person from whom it came, and because it is unlikely to be sensitive material, most participants would be less concerned about it being kept confidential. However, this doesn't negate the need for care: data sets may contain or be easily linked to names and other personal details if they are not effectively anonymised or if the personal details of participants and the data are kept in too close a proximity to one another--in the same file or folder for example. And it shouldn't be assumed that data the researcher considers altogether non-sensitive is non-sensitive from all the participants' points of view.
Moreover, physiotherapists may well carry out qualitative research. In qualitative research, identifying details may well be an integral part of the data--for example, names, places, dates in a taped interview narrative. Anyone who happens to have access to this data may be able to identify the participant or others in the narrative. Those who have such access may include people other than the researcher--for example a person who transcribes the interview. Confidentiality may also be breached through dissemination of results. Although information such as name, age and location will be expected to be removed prior to publication, there is always potential for participants to be recognised through other identifying information. This is a particular concern in a small country such as New Zealand where including data in a published article about a participant such as a diagnosis, the number of children, and that the participant lives in a small town may be sufficient to result in identification of a participant. This problem may be avoided by changing personal and other details.
In responding to concerns about confidentiality and privacy, researchers generally make certain promises. Typically promises about confidentiality include guarantees to keep data gathered for research purposes in a locked filing cabinet in a locked cupboard or office, or on a password-protected computer. However, these promises may presuppose ideal conditions, rather than the reality of practice. One can easily imagine circumstances in which a busy physiotherapist comes into his/her office, putting things down momentarily on a desk, only to be called away to take a phone call or see a client who is in a hurry. Anyone who happens to enter the office may see the things dumped on the desk, including photos of someone known to him/her, and suddenly good intentions about confidentiality fall down.
These realities of practice aren't unique to physiotherapists: however, they remind physiotherapy researchers that confidentiality in practice is achieved only with constant vigilance; that it involves many practicalities, and that the real contexts in which people exist often need to be carefully monitored to fulfil the promise of confidentiality.
Non-maleficence and beneficence: Ensuring the well-being of participants
As stated earlier, there is a need to consider the harms and benefits that might eventuate from research. As far as doing harm goes, it may be considered that physiotherapy is unlikely to be ethically troubling. It is fairly clear what sorts of harms may arise from taking substances into the body in the form of drugs or from new surgical techniques: both are in their different ways physically quite invasive. But it might be argued that physiotherapy is not anywhere near so physically invasive, and doesn't have the toxic potential of the drug or the serious injury potential of the scalpel. This is potentially a problematic attitude, and there are at least three strong reasons for taking the potential for harms eventuating from physiotherapeutic interventions being researched very seriously.
The first reason is that some physiotherapy interventions do contain serious potential for physical harm. Electrotherapy modalities, acupuncture spinal manipulation and even some forms of mobilization can all result in undesirable outcomes. Therefore we need to consider any potential for physical harm when involving patients in research even on well accepted procedures.
The second reason is that potential for harm may result from the adoption of new treatments, or the adaptation of old treatments on populations in which such treatments have not previously been used. In 1998, the Journal of Pediatrics published an article written by Harding, Miles, Becroft, Allen and Knight. The article concluded that encephaloclastic porencephaly--a characteristic form of brain damage, sometimes with fatal or severely disabling outcomes--might be a previously unrecognized complication of chest physiotherapy in vulnerable extremely preterm infants. The article centred on a series of cases in Auckland. The outcomes in Auckland were reminiscent of those in an earlier article, written by Cross et al (1992), which reported on a case series in Birmingham, in the UK. In both cases, the physiotherapy that was being given was an established treatment; however it had not been tried before on very premature babies. It was extended to their care presumably because it seemed to have advantages in other age groups, but in retrospect physiotherapy emerged as a possible cause of serious, sometimes mortal, outcomes.
The Birmingham and Auckland case series alert us to a number of important things about physiotherapy research. First, that research needs to be done--innovative extensions of normal treatments require careful monitoring and checking not only retrospectively, but prospectively. Second, that innovative physiotherapy is potentially extremely harmful. This may be particularly the case where the patient group which it is thought might benefit from the physiotherapy is itself physically vulnerable.
The third reason is because other sorts of harm can arise from the non-physical research carried out by physiotherapists, which might be perceived as very low-risk methods such as interviewing. For example, the recollection of emotionally difficult experiences in the course of an interview can cause harm in participants. It is important that this possibility is identified. Recognising the possibility of such emotional and/or psychological harm is the first step to putting in place safeguards to prevent such harm. If physiotherapy researchers themselves do not recognise the potential for this form of harm, they may not recognise that ethical review of their study is required through an ethics committee.
Further responsibilities to participants
There are times during research when a researcher may come across information that has significant implications for the health or wellbeing of the participant. For example, a physiotherapist may carry out research on elderly people, and while screening participants for research into falls may find out a participant has very high blood pressure. A question arises about what physiotherapists should do with this kind of information. Should they report this concern, or keep such findings confidential? If the participant is not a patient of the physiotherapist, the physiotherapist may feel that he or she has no obligation to respond at all. However a duty of care to participants may indeed extend further than the described limits of the research. To some extent these concerns should have been predicted prior to commencement of the study and appropriate steps put in place of which the participant is made aware as a part of the consent process. Such steps might include ...
A further example is when undertaking research on dependent people, a researcher may be told during an interview or simply in passing that a participant is suffering abuse at the hands of a caregiver. In this example, a researcher may be unsure about exactly how to respond. They might be fearful that they may make the situation worse if they respond inappropriately. There is also a possibility that such information is not true and in passing it on, the physiotherapist risks impugning the name of the caregiver. A physiotherapist may need to access other health care services in order to respond in a sensitive manner. Appropriate mechanisms for responding to such possibilities should be provided when submitting an application for ethical committee approval.
Our primary aim has been to emphasise three things with respect to ethical conduct of physiotherapy research. First, that independent ethical review of research is vital, but that it is not a substitute for the researcher's own ethical sensitivity and awareness. Second, that the ethical ideas which underlie the traditional foci of research ethics (for example the idea of autonomy which underlies informed consent and confidentiality) need to be supplemented by taking into account the implications of other ethical ideas, such as the avoidance of harm and doing good. And third, that ethical conduct of physiotherapy research requires a focus on the realities of the research context, and on the details of what can happen, on the part of the researcher.
* Where possible, physiotherapy treatment should be based on research conducted under the widely accepted values of bioethics
* In practical situations there are many hidden threats to the effective application of these values
* Physiotherapists in New Zealand are expected to seek independent ethical review of any proposed research
Beauchamp T, Childress . (2001). Principles of medical ethics. Fifth Edition. New York: Oxford University Press Beecher H (1966). Ethics in Clinical Research. New England Journal of Medicine 274(2); 1354-1360.
Campbell A, Gillett, G, and Jones, G. (2005) Medical Ethics. Oxford: Oxford University Press.
Cartwright S (1988). Cartwright Report: Committee of Inquiry into allegations concerning the treatment of cervical cancer at National Women's Hospital and into other related matters. Auckland: Government Printing Office.
Coney S Bunkle P (1986). An 'unfortunate experiment' at National Women's. Metro, 46-65.
Cross J, Harrison C, Preston D, Rushton D, Newell S, Morgan, M, Durbin G (1992). Postnatal encephaloclastic porencephaly--a new lesion? Archives of Disease in Childhood. 67: 307-311.
Duncan A, Dunstan G and Welbourn R (1981) Dictionary of Medical Ethics. London: Darton, Longman & Todd. pp210-211 Evans D, Evans M (1996) A decent proposal. John Wiley and Sons
Harding J, Miles D, Becroft D, Allen B, Knight D (1998); Chest physiotherapy may be associated with brain damage in extremely premature infants. Journal of Pediatrics 132(3): 440-444.
Health & Disability Commissioner (1996). Code of Health and Disability Services Consumers' Rights. http://www.hdc.org. nz/theact/theact-thecode [Accessed January 29, 2008].
Health Research Council of New Zealand (2007). Guidelines for Health Research with Children.
Health Research Council of New Zealand. (2008). Guidelines for Researchers on Health Research Involving Maori.
Henley L, Frank D (2006) Reporting Ethical Protections in Physical Therapy Research. Physical Therapy 86(4); 499-509.
Ministry of Health (1994). Health Information Privacy Code. Wellington: Ministry of Health New Zealand.
Ministry of Health (2006) Operational Standard for Ethics Committees.
National Ethics Advisory Committee (2006). Ethical Guidelines for Observational Studies: Observational Research, Audits, and Related Activities.
National Ethics Advisory Committee. (2008). Work Programme. Available at http://www.neac.health.govt.nz/moh.nsf/ indexcm/neac-workprogramme (accessed 30 May 2008).
Nie J-B, Anderson L (2003) Bioethics in New Zealand: A historical and sociological review. In John Peppin and Mark Cherry, (Eds) Annals of Bioethics. The Netherlands: Swetz Zeitlinger, pp. 341-361.
Nuremberg Code (1946-1949). Nuremberg Code. In: Trials of war criminals before the Nuremberg military tribunals under control council law. No.10. Washington, D.C: United States Government Printing Office.
Paul C (1988). The New Zealand cervical cancer study: could it happen again? British Medical Journal 297(6647):533-9.
Sporle A and Koea J (2004) New Zealand Medical Journal 117, 1199
Tolich, M & Baldwin, K. (2005) Informing consent in New Zealand research: researchers' conflict of interest and patient vulnerability. New Zealand Medical Journal 118(1210).
World Medical Association: Declaration of Helsinki: (2000) Ethical Principles for Medical Research Involving Human Subjects.; Section B(13).
Lynley Anderson, DipPhysio, MHealSc, PhD
Neil Pickering, BA(Hons), MA, PhD
Senior Lecturers, Bioethics Centre, Medical and Surgical Sciences, Dunedin School of Medicine
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Scholarly Paper|
|Author:||Anderson, Lynley; Pickering, Neil|
|Publication:||New Zealand Journal of Physiotherapy|
|Date:||Nov 1, 2008|
|Previous Article:||Physiotherapy and the new disciplinary process.|
|Next Article:||Mulligan's mobilisation with movement: a review of the tenets and prescription of MWMs.|