Annals of Oncology Advance Access originally published online on January 17, 2006
Annals of Oncology 2006 17(4):702-708; doi:10.1093/annonc/mdj136
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 European Society for Medical Oncology
Do oncologists discuss expensive anti-cancer drugs with their patients?
1 Division of Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia; 2 Oxford Uehiro Centre for Practical Ethics, University of Oxford, Oxford, UK
* Correspondence to: Dr M. Jefford, Division of Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Locked Bag 1, A'Beckett Street, Victoria 8006, Australia. Tel: +61 3 9656 1697; Fax: +61 3 9656 1408; E-mail: Michael.Jefford{at}petermac.org
 |
Abstract |
|---|
 Top Abstract introductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
Background: In Australia, some anti-cancer drugs are only available at significant financial cost to patients. We sought the views and practices of Australian medical oncologists regarding discussion of high cost drugs (HCDs).
Participants and Methods: A postal survey was mailed to all 274 members of the Medical Oncology Group of Australia. Three clinical scenarios described HCDs associated with either improved overall survival, encouraging response rate in a treatment-refractory cancer, or a scenario with improved treatment tolerability. Participants were asked about their discussion and prescription of HCDs.
Results: There was a 78% response rate. Most respondents were male (71%), worked in a metropolitan practice (87%) and spent more than 50% of their working time in patient care (87%). Forty-eight percent had previously prescribed a HCD. In the three scenarios, respondents would generally prescribe the drug if it were subsidised, however, between 28% and 41% (depending on the scenario) would not mention the HCD if it were not subsidised. Major reasons for not mentioning the HCD were concerns that discussion would ‘worry the patient’ or that the doctor would ‘feel bad’.
Conclusions: Despite literature suggesting that patients wish to be well informed and active participants in decision making, the practice of a significant percentage of Australian medical oncologists may prevent this.
Key words: chemotherapy, decision making, expensive drugs, medical ethics, patient participation
|
introduction |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
Patients generally wish to receive as much information as possible, particularly in relation to treatment options [1
, 2], and wish to be active participants in treatment decision-making [1, 3], yet cannot make informed decisions unless they are presented with all relevant information [4]. Shared decision-making is considered the best practice approach to negotiating treatment plans [5], both according to professional practice guidelines [6] and the law [7] and is the approach preferred by most patients [1].
In Australia, many of the newest anti-cancer drugs, with proven efficacy in phase II and III trials, face delays in approval or subsidisation by the Pharmaceutical Benefits Scheme (PBS). Drugs are included on the PBS following advice from the Pharmaceutical Benefits Advisory Committee (PBAC). The PBAC considers a number of issues before recommending that a drug be listed on the PBS. These include efficacy and quality of life benefits, safety, cost and the overall PBS budget. Some drugs are not subsidised because of overall cost constraints, even though they might represent preferred therapeutic options, compared with subsidised drugs. If a patient wishes to receive an unsubsidised drug, they are required to fund the full cost of the drug themselves. Costs can be considerable – commonly around A$1000 per week (
600, £420, US$730). For most people this would represent a considerable financial burden. For the purpose of this study, medications in this category are referred to as high-cost drugs (HCDs).
Informing patients about unsubsidised HCDs raises a number of ethical issues for the individual clinician [8]. These include dilemmas about what is in the patient's overall best interests, perhaps justified paternalism; concern for equity and justice, and most importantly how to respect personal autonomy.
The primary aim of this study was to document the attitudes and practices of Australian medical oncologists with regard to disclosure of HCD treatment options to patients, using three clinical scenarios based on recent clinical trial outcomes.
|
methods |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
procedure
A covering letter, questionnaire and reply-paid envelope were mailed to all 274 members of the Medical Oncology Group of Australia (MOGA, the representative body for medical oncologists in Australia) in April 2003. Confidentiality was maintained by the use of a unique identification code. Of the 274 members, 65 were listed as medical oncologists in training. Telephone follow-up was conducted to enhance the response rate. Up to three attempts were made to contact non-responders.
the questionnaire
The questionnaire utilised three case scenarios, each involving the potential use of a HCD for a patient with advanced cancer with a good performance status. Each scenario was modelled on actual HCDs not subsidised by the PBS at the time the questionnaire was developed. The three scenarios were:
- scenario A: A HCD shown in a phase III study involving people with metastatic cancer to result in an improvement of median overall survival of five months (median overall survival extended from around 24 to 29 months). This was based on a study of trastuzumab in combination with chemotherapy for women with metastatic breast cancer [9].
- scenario B: A HCD shown in a phase II study (no phase III data available) involving patients with a malignancy resistant to all other forms of treatment, to result in a high objective tumour response. This scenario was based on imatinib in gastrointestinal stromal tumours (GIST) [10
].
- scenario C: A HCD shown in a phase III study to result in similar response and survival benefit as standard chemotherapy, but with significantly improved patient tolerability – based on a study of gemcitabine with cisplatin versus ‘MVAC’ (methotrexate, vinblastine, doxorubicin, and cisplatin) combination chemotherapy for patients with metastatic bladder cancer [11
].
- scenario B: A HCD shown in a phase II study (no phase III data available) involving patients with a malignancy resistant to all other forms of treatment, to result in a high objective tumour response. This scenario was based on imatinib in gastrointestinal stromal tumours (GIST) [10
600, £420, US$730). |
measures |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
demographic details
Included age, sex, number of years of oncology practice, location of practice (metropolitan/regional) and percentage of the working week spent in direct patient care.
self-reported practices
Respondents were asked whether or not they would be prepared to prescribe a HCD if a patient was prepared to pay for it, and whether they had personal experience of doing so.
decision-making style
They were also asked to indicate their overall treatment decision-making style, which was adapted from an existing scale [12].
attitudes to prescribing HCDs
Items to assess doctor's attitudes in relation to each of the three scenarios were developed by a panel of experts with extensive experience in doctor–patient communication and clinical decision making. These measures were pretested by 10 medical oncologists. Individual items included:
- If the HCD were funded through the PBS or by the hospital, would you prescribe it in this clinical scenario?
- If it were not available, would you inform the patient about the HCD, even though the patient or family would need to finance the entire cost of the drug?
- If you would not inform the patient, what is/are your reasons for not doing so? [A list of options was provided (see Figure 1) and the respondent was asked to indicate all possible reasons and also to nominate their single most important reason for not informing the patient of the HCD.]
- Which of the following circumstances would change your decision? [A series of options was provided (see Figure 2) and the respondent was asked to indicate all possible reasons].
- If you would tell the patient about the HCD, how would you inform the patient? [The respondent was asked to specify one from a series of options; see Figure 3].
|
|
|
statistics
Descriptive statistics were used to describe the frequency of responses. Chi-squared tests were used to determine associations between doctor characteristics and self-reported attitudes and practices regarding discussion of HCDs. The level of significance was set at a P value of 0.05.
|
results |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
demographics/decision-making preferences
Of the 274 MOGA members, 38 were ineligible (four on extended leave or overseas, three on maternity leave, four no longer saw patients and 27 were not able to be located). Of the remaining 236, 184 returned questionnaires (response rate 78%).
Most respondents were male (71%), worked in metropolitan practice (87%) and spent more than 50% of their working time in patient care (87%). Forty-six percent were aged 40 or under and 33% were aged 41–50. Twenty-three percent had been practising in medical oncology for five years or less and 21% had been practising over 20 years. This profile is typical of the whole MOGA membership and suggests the sample is representative.
In relation to self-reported decision-making styles, most reported that they prefer to make treatment decisions, but strongly consider the patient's needs and priorities (42%), or indicated that they prefer to make decisions with the patient on an equal basis (35%). Of the remainder, 4% preferred the patient to make decisions and 1% preferred to make decisions alone. No significant associations were seen between self-reported decision-making styles and gender, time in medical oncology practice and location of practice. However, oncologists in training were significantly less likely to report a preference for patient-focussed decision making (P = 0.02).
previous prescription of HCDs
The vast majority of respondents (95%) indicated that they would prescribe a HCD if they knew the patient was prepared to pay for it, and 48% indicated that they had already done so. Respondents aged 40 and under, those practising in oncology for five years or less and advanced trainees were significantly less likely to have previously prescribed HCDs (P < 0.001 for all). Males were significantly more likely to have prescribed HCDs than females (58% versus 25%, P < 0.001). Location of practice did not appear to influence the likelihood of having previously prescribed a HCD.
informing patients about HCDs
If the patient did not have to pay for the HCD, 92% of respondents would be prepared to prescribe the drug in scenario A, 72% in scenario B and 94% in scenario C. The main reasons why oncologists would not prescribe the HCD described in scenario B were: inadequate evidence and/or the need for phase III data (57%); lack of survival and toxicity data (33%), and cost/lack of cost–benefit analysis (10%). Some respondents gave more than one reason. Even though the scenario was based on a genuine clinical trial, three respondents (6%) stated that the scenario was ‘too good to be true’.
If the patient would have to pay for the HCD themselves a significant number of oncologists would not inform the patient about the HCD (scenario A 34%, B 41%, C 28%). Reasons for not informing patients about a HCD are shown in Figure 1.
For scenario A, the most common reasons for not informing the patient were: ‘the knowledge that they could not obtain this new drug would be too distressing for the patient and their family’ (endorsed by 92% of respondents and most frequently selected as the most important reason), ‘I would feel bad mentioning a drug that the patient probably cannot afford’ (83%), and ‘the knowledge that some patients can afford to pay for the HCD will only cause distress in those who cannot’ (65%).
In scenario B, the most common reasons were: ‘there is not enough evidence available on the efficacy of the new drug’ (92%, and the single most important reason), ‘the knowledge that they could not obtain this new drug would be too distressing for the patient and their family’ (49%) and ‘costs to the patient outweigh the benefit’ (46%).
In scenario C, the most common reasons were: ‘I would feel bad mentioning a medication that the patient probably cannot afford’ (77%), ‘the knowledge that they could not obtain this new drug would be too distressing for the patient and their family’ (67%, this was the most frequently noted single most important reason) and ‘costs to the patient outweigh the benefit to the patient’ (54%).
The least favoured explanation for not discussing HCDs in each of the scenarios was ‘it is too time consuming to discuss all the potential treatment options’ (8%, 9% and 4%).
circumstances that might change a decision to not discuss a HCD
Figure 2 shows the frequency of respondents changing away from a decision not to discuss a HCD treatment. In scenarios A and B factors that might change an intention to not discuss a HCD included: ‘the patient or family want to know all possible treatment options’ (90% for scenario A, 65% B), ‘the patient is fully privately insured, assuming the insurer will pay for the drug’ (83% for A, 54% for B) and ‘the drug cost is A$5000, rather than A$30 000 or A$25 000’ (82% for A, 54% for B). For scenario C, the chief reasons were: ‘the drug cost is A$5000, rather than A$20 000’ (77%), ‘the patient or family want to know all possible treatment options’ (76%) and ‘the patient is fully privately insured, assuming the insurer will pay for the drug’ (74%). The least frequently endorsed reason to change a decision regarding discussion of a HCD treatment in both scenarios A and B was ‘the patient is an important figure in the community’ (13%, 13%), and for scenario C ‘the patient is 25 (rather than 59) years old’ and ‘the patient has stated that they want to live as long as possible’ (11%, 11%).
how the HCD would be discussed
For those who would discuss the HCD, figure 3 indicates the ways the HCD could be discussed. For all scenarios, the majority of respondents indicated that they would introduce the new treatment as an option, but indicate the costs involved (73% for A, 68% for B and 56% for C). Very few respondents stated they would introduce the HCD as the preferred alternative to standard treatment and ask the patient to obtain their own funding (0% for A, 4% for B and 2% for C).
|
discussion |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
In this study, we sought the opinions and practices of Australian medical oncologists regarding discussion of HCDs. We presented three clinical scenarios in which, compared with standard therapy, treatment with a HCD was associated with a significant, objective patient benefit. The majority of oncologists indicated that they would discuss the HCD with the patient if it were subsidised. Interestingly, nearly 30% of practitioners would not inform patients of treatment B even if it were subsidised by the PBS. This scenario described results similar to those from a phase II study of imatinib for GIST [10
]. This raises an interesting question about when it can be said that there is enough evidence or confidence to support the use of a new anti-cancer agent. Nevertheless, the major finding from this study was that, if the drugs were not subsidised, oncologists were less likely to discuss the HCD option. Frequently given reasons for not discussing the HCD were ‘the knowledge that they could not obtain this new drug would be too distressing for the patient and their family’ and ‘I would feel bad mentioning a medication that the patient probably cannot afford’ (figure 1). This indicates that oncologists were concerned about the potential emotional impact this discussion might have upon the patient and their family and acknowledged that these discussions are stressful for practitioners. Oncologists were not swayed to discuss a HCD by virtue of the patient's age, family circumstance or social standing, but were swayed if they felt that the patient could afford or might be inclined to pay for the treatment, or if the patient was explicit in wanting to know all possible treatment options (Figure 2).
Other countries (for example, the United Kingdom) share a similar two-stage system of drug approval and subsequent subsidisation, as in Australia. They also encounter situations where a new drug may be approved, but not subsidised. We are uncertain whether the findings of our survey might apply in these countries and whether the findings might apply in countries that have quite a different model of health care delivery (for example, in the United States).
patients desire for information
Numerous studies have shown that patients with cancer generally wish to be fully informed [1]. However, the question of whether patients with cancer want to be informed about HCDs has not been specifically studied. The Australian High Court, through the landmark cases of Rogers versus Whitaker [7] and Chappel versus Hart [13], has emphasised the importance of providing patients with any and all information that may be of importance to their personal decision-making. Clinicians are known to be poor judges of their patients' preferences for information [14, 15]. Our study indicates that oncologists are often gatekeepers of information. Depending on the clinical scenario, between 28 and 41% of oncologists would not discuss the option of a HCD with their patients. Oncologists appear to be making assumptions about what patients want to know and how much patients would or should spend on their treatment. Though we did not ask whether they would attempt to elicit information from the patient to confirm or refute these assumptions, it seems that paternalistic assumptions may influence treatment practice in relation to HCDs. Elsewhere, we have contributed an extensive ethical analysis of this scenario – discussing expensive, unfunded treatment options [8].
do patients wish to participate in treatment decisions?
Historically, physicians have assumed the dominant role in treatment decision-making. However, there has been a growing emphasis on patient autonomy and active involvement in shared decision making [3, 16–19]. As with information preferences, doctors are often unable to predict patient preferences regarding involvement in treatment decision making. It is recommended that health care providers ascertain the extent to which their patients desire to participate in treatment decisions [21, 21]. In our study most oncologists indicated a preference for shared decision making. However, it is unclear what shared decision making means if patients are not informed of all possible treatment options. It should be noted that the validity of doctors' self-reporting of decision-making style has not been established; hence, ratings may not accurately reflect actual practice with patients.
Interestingly, we found that medical oncology trainees were more likely to prefer a style of decision making that is less patient-focussed. Our data do not provide clues as to why trainees demonstrate this preference. Perhaps, while learning, it is easier to conform to a traditional, medically dominant role. Perhaps too, such discussions are difficult, both for clinician and patient and are most avoided by those who are least experienced.
how should oncologists discuss HCDs with their patients?
Our survey indicates that medical oncologists in Australia often fail to discuss the option of HCDs with their patients and that oncologists find these discussions difficult. A frequently cited reason for not mentioning a HCD option focussed on the distress that this conversation would engender in the oncologist (as well as the patient). Yet oncologists routinely deliver bad news, such as the news of cancer progression. From a patient perspective, in addition to the stress of a cancer diagnosis, there are many other obstacles to effective communication and truly shared decision making. These include an unfamiliar language, descriptions of risk and unfamiliar concepts, such as median survival [22]. Yet these ideas are central to discussions regarding chemotherapy drugs. Guidelines provide advice for health professionals to promote patient understanding and to facilitate shared decision making [6]. Further guidelines could be developed to assist clinicians with the difficult task of discussing HCDs. A guide has been recommended for the discussion of cancer clinical trials, which emphasises discussion of standard therapies, prior to discussing the possibility of participation in a clinical trial [23]. A similar sequence could be recommended, involving discussion of standard options, before discussing the possibility of a HCD. Oncologists who do mention HCD options appear to use a similar model, introducing the HCD as an option, but indicating the costs involved (Figure 3).
|
conclusions |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
A significant percentage of medical oncologists do not advise their patients of HCD options. They refrain from these discussions primarily because of concern for their patients, but also because they recognise that it is personally challenging to have these discussions. However, most patients express a desire to be fully informed and to be active participants in treatment decision-making. We have argued elsewhere that it is unethical and paternalistic to withhold such information from patients [8
]. Doctors are not in a position to decide what their patients should know or how they should spend their resources. A major challenge for clinical practice is how best to conduct these potentially difficult discussions. We suggest that communication skills training may be an important part of this process. It is unlikely that the issue of unfunded, expensive anti-cancer drugs will soon go away. Indeed, as the cost of drug development rises, this is likely to be a significant issue for all medical practice. |
Acknowledgements |
|---|
We thank staff from The Cancer Council Victoria, who assisted with telephone follow-up of surveyed oncologists.
Received for publication October 5, 2005. Revision received December 4, 2005. Accepted for publication December 7, 2005.
|
References |
|---|
|
TopAbstractintroductionmethodsmeasuresresultsdiscussionconclusionsReferences |
|---|
1. Jefford M, Tattersall MH. Informing and involving cancer patients in their own care. Lancet Oncol 2002; 3: 629–637.[CrossRef][Web of Science][Medline]
2. Jenkins V, Fallowfield L, Saul J. Information needs of patients with cancer: results from a large study in UK cancer centres. Br J Cancer 2001; 84: 48–51.[CrossRef][Web of Science][Medline]
3. Laine C, Davidoff F. Patient-centered medicine. A professional evolution. J Am Med Assoc 1996; 275: 152–156.
4. Sowden AJ, Forbes C, Entwistle V, Watt I. Informing, communicating and sharing decisions with people who have cancer. Qualit Health Care 2001; 10: 193–196.
5. Brock DW. The ideal of shared decision making between physicians and patients. Kennedy Inst Ethics J 1991; 1: 28–47.[Medline]
6. National Breast Cancer Centre and National Cancer Control Initiative. 2003. Clinical practice guidelines for the psychosocial care of adults with cancer. National Breast Cancer Centre, Camperdown, NSW.
7. Rogers vs Whitaker (1992) 175 CLR 479.
8. Jefford M, Savulescu J, Thomson J et al. Medical paternalism and the practice of discussing high-cost unsubsidised drugs with patients. Brit Med J 2005; 331: 1075–1077.
9. Slamon DJ, Leyland-Jones B, Shak S et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001; 344: 783–792.
10. Joensuu H, Roberts PJ, Sarlomo-Rikala M et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001; 344: 1052–1056.
11. von der Maase H, Hansen SW, Roberts JT et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000; 18: 3068–3077.
12. Cassileth BR, Zupkis RV, Sutton-Smith K, March V. Information and participation preferences among cancer patients. Ann Intern Med 1980; 92: 832–836.[CrossRef][Web of Science][Medline]
13. Chappel vs Hart [1998] HCA 55.
14. Degner LF, Kristjanson LJ, Bowman D et al. Information needs and decisional preferences in women with breast cancer. J Am Med Assoc 1997; 277: 1485–1492.
15. Fallowfield L, Ford S, Lewis S. No news is not good news: information preferences of patients with cancer. Psychooncology 1995; 4: 197–202.[Medline]
16. Bekker H, Thornton JG, Airey CM et al. Informed decision making: an annotated bibliography and systematic review. Health Technol Assess 1999; 3: 1–156.[Medline]
17. Robinson A, Thomson R. Variability in patient preferences for participating in medical decision making: implication for the use of decision support tools. Qualit Health Care 2001; 10 Suppl 1: i34–38.
18. Charles C, Gafni A, Whelan T. Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango). Soc Sci Med 1997; 44: 681–692.[CrossRef][Web of Science][Medline]
19. Charles C, Whelan T, Gafni A. What do we mean by partnership in making decisions about treatment? Brit Med J 1999; 319: 780–782.
20. Fallowfield L. Participation of patients in decisions about treatment for cancer. Brit Med J 2001; 323: 1144.
21. Harris KA. The informational needs of patients with cancer and their families. Cancer Pract 1998; 6: 39–46.[CrossRef][Web of Science][Medline]
22. Bunker JP, Houghton J, Baum M. Putting the risk of breast cancer in perspective. Brit Med J 1998; 317: 1307–1309.
23. Brown RF, Butow PN, Butt DG et al. Developing ethical strategies to assist oncologists in seeking informed consent to cancer clinical trials. Soc Sci Med 2004; 58: 379–390.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Mileshkin, P. E. Schofield, M. Jefford, E. Agalianos, M. Levine, A. Herschtal, J. Savulescu, J. A. Thomson, and J. R. Zalcberg To Tell or Not to Tell: The Community Wants to Know About Expensive Anticancer Drugs As a Potential Treatment Option J. Clin. Oncol., December 1, 2009; 27(34): 5830 - 5837. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Jefford, L. Mileshkin, P. Schofield, E. Agalianos, J. Thomson, and J. Zalcberg Discussing Expensive Anticancer Drugs J. Clin. Oncol., January 20, 2009; 27(3): 476 - 477. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||