ReviewAre we there yet? The state of the evidence base for guidelines on breaking bad news to cancer patients
Introduction
Building an evidence base is an integral part of ethical care.1 Level 1 evidence, determined by multiple randomised controlled trials reporting similar results, is accepted as the most rigorous level of evidence.2 In general, there is an expectation that research efforts will achieve higher levels of evidence over time.3, 4
The provision of psychosocial care is a new-comer to cancer patient care,5 therefore Level 1 evidence for psychosocial care may take time to emerge. This may be due to a low likelihood of harm as a result of psychosocial intervention and a lack of financial incentives for psychosocial trials. Nevertheless, Level 1 evidence is still required in order to provide best-evidence psychosocial care.
Over their career cancer oncologists will break bad news an average of 20,000 times.6, 7 However, many clinicians have difficulty or lack the necessary skills in providing this type of care.8, 9 Data gathered from studies examining interactions between patients and clinicians suggest that the consequences of breaking bad news poorly can affect both the patient10 and the doctor involved.11 For the doctor, stress related to these incidents can contribute to burn-out and anxiety.12 For the patient, poorly communicated bad news can result in increased stress and anxiety, poor adjustment to the bad news, and poorer health outcomes.10, 13, 14, 15
Following recognition of the importance of the issue and the lack of evidence on how to effectively break bad news to patients, a consensus approach was taken to develop practice guidelines.16, 17 The consensus guidelines recommended that the evidence base for breaking bad news be developed via rigorous intervention studies.17 In the absence of an evidence base, it is possible that provider education in breaking bad news may be idiosyncratic or misguided. This could result in practices which are ineffective or even detrimental to patients’ psychosocial health.
The first requirement to advance the evidence base in breaking bad news to cancer patients is new data. While consensus views may be based on expert opinion, an evidence base requires empirical research. Second, there is a need for intervention studies which describe and test strategies for improving the communication of bad news. Reviews in non-cancer healthcare fields revealed that the majority of studies had a descriptive focus.18 Similarly, a 2004 Cochrane review exploring communication skills training for healthcare professionals working with cancer patients found only three intervention studies related to breaking bad news.19
A third characteristic crucial to advancing the evidence base in breaking bad news is demonstration of an improvement in psychosocial outcomesfor patients. Guidelines have been based on an understanding of the negative effects of the way in which a clinician breaks bad news (i.e. anxiety, distress or dissatisfaction of the patient). However, the early availability of guidelines5 and educational courses20 may have focused researchers on skills rather than on patient benefit. Although improved skills are assumed to improve psychosocial outcomes, this association must be demonstrated in order to justify the resources required to provide training and achieve sustained improvements. Examining psychosocial impacts on providers should be contingent on establishing that it is possible to achieve psychosocial benefits for patients.
Reviews of breaking bad news literature in other health fields have identified a fourth key characteristic of research which will advance the evidence base – design rigour. These reviews found that only a minority of the identified interventions met recognised criteria for good research design and methodological quality.21 Achieving methodological rigour is vital because the quality of the methodology determines the quality of the evidence.5 A lack of research rigour in a field may lead to conflicting results or even systematic error.
Therefore, there is a need to examine the research on breaking bad news to identify whether efforts have been directed towards achieving improved outcomes for patients.19, 22 The aims of this review were to examine the literature on breaking bad news to cancer patients published since 1995 to identify:
- (1)
The proportion of studies which provided new data and therefore potentially added to knowledge in the field.
- (2)
The proportion of studies providing new data which tested the effectiveness of intervention strategies for improving breaking bad news.
- (3)
The proportion of intervention studies which included a patient psychosocial outcome as one of the study measures. Psychosocial outcomes included anxiety, depression, stress, satisfaction and quality of life.
- (4)
The proportion of the intervention studies which measured psychosocial outcomes for patients and met accepted criteria for design rigour.
- (5)
Whether a positive effect on patients’ psychosocial outcomes was identified as a result of rigorous interventions to improve the communication of breaking bad news.
Section snippets
Search strategy
A computer-based literature search was performed to identify publications on the topic of breaking bad news to cancer patients, published between January 1995 and March 2009. The year 1995 was chosen as the search start date as the consensus guidelines for breaking bad news were published in 1995.17 Medline and PsycINFO databases were chosen for the search as both contain research publications across a wide range of health professions in the field of oncology, and are leading sources for
Publication search and exclusion
A total of 443 publications were identified. One hundred and ninety-eight (44.7%) publications were not relevant as they: were duplicates; did not involve humans; contained non-cancer disease-specific prevalence data; were book chapters or texts; or were dissertations or conference papers. Three papers23, 24, 25 provided additional outcome data on previously reported interventions.26, 27 Multiple papers on the same intervention with the same sample were “counted” as one study.
New data
Two-hundred and
Discussion
The review revealed that an extremely small proportion (1.6%) of the published studies in the field of breaking bad news involved studies which provided an evidence base for practice. Of all publications addressing the issue of breaking bad news to cancer patients, approximately half provided new data and less than one-fifth involved an intervention trial. More than 90% of the intervention effort failed to measure psychosocial outcomes for patients. This is despite the published observation in
Conflict of interest statement
None declared.
Acknowledgements
The authors would like to thank the Hunter Medical Research Institute (HMRI) for providing the infrastructure funding that made this paper possible. The support of Miss Aimee Mavratzakis in coding the papers is also acknowledged.
References (30)
- et al.
Recipients’ perspective on breaking bad news: how you put it really makes a difference
Patient Educ Couns
(2005) - et al.
Ability of primary care physicians to break bad news: a performance based assessment of an educational intervention
Patient Educ Couns
(2006) - et al.
We are what we do: research outputs of public health
Am J Prev Med
(2008) - et al.
Factors that influence cancer patients’ anxiety following a medical consultation: impact of a communication skills training programme for physicians
Ann Oncol
(2006) - et al.
Publication bias in clinical research
Lancet
(1991) - Evidence-Based Medicine Working Group. Evidence-based medicine: a new approach to teaching the practice of medicine....
- US Preventive Services Task Force. Guide to clinical preventive services. 2nd ed. Baltimore: Williams and Wilkins;...
- et al.
Health research in Australia: its role in achieving the goals and targets
Health Promot J Austr
(1994) - et al.
Behavioural alcohol research: new directions or more of the same?
Addiction
(1997) - et al.
Breaking bad news 1: current best advice for clinicians
Behav Med
(1998)