|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 10 (April 1), 2008: pp. 1724-1731 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.13.5566 Physician-Identified Factors Affecting Patient Participation in Reaching Treatment Decisions
From the Medical Psychology Research Unit, Department of Cancer Medicine, and School of Psychology, University of Sydney, Sydney, Australia Corresponding author: Heather Shepherd, Medical Psychology Research Unit, Blackburn Building (D06), University of Sydney, New South Wales 2006, Australia; e-mail: heathers{at}med.usyd.edu.au
Purpose Cancer physicians report high comfort with shared decision making but a lower frequency of using this approach in practice. Information regarding physicians’ perceptions of what helps and what hinders patient involvement in decision making may facilitate understanding of this discrepancy. Methods We surveyed 604 Australian cancer physicians treating breast, colorectal, gynecologic, hematologic, or urologic cancer to investigate barriers and facilitators to reaching treatment decisions with their patients and their support of strategies to encourage patient involvement and reflection on treatment options. Factor analysis and regression analyses were used to investigate relationships between variables and identify predictors of greater reporting of barriers to sharing treatment decisions with patients. Results Insufficient information at the first consultation (28.9%) and insufficient time (28.4%) were the most frequently reported barriers to reaching treatment decisions with patients. Multivariate analysis revealed that less experienced physicians more commonly reported system barriers (P = .00). Patients trusting their physician and being accompanied at the consultation were most helpful to reaching a treatment decision. Providing written information about treatment options, making a further appointment to reach a decision, encouraging the patient to speak with their family physician and treatment team, and the presence of a third person during the consultation were felt to encourage involvement and reflection on treatment decisions. Conclusion Cancer physicians experience difficulties when reaching treatment decisions with their patients. Interventions and strategies that physicians support are required to enhance patient involvement in reaching a treatment decision.
Involving patients in discussion about treatment options and reaching a treatment decision is increasingly common. Cancer patients report wanting to be involved in treatment decision making; however, this preference is often not achieved when consulting with their cancer physicians.1-4 Public expectation to be fully informed about health care and available interventions has increased over recent years, alongside decreasing acceptance of medical paternalism.5 Patients who are involved in reaching a treatment decision report increased satisfaction, reduced decisional conflict, and improved compliance with treatment.6,7 We surveyed cancer clinicians across Australia from August 2004 to May 2006 to document their comfort with and use of different decision-making approaches and their attitudes toward treatment decision making with patients. We found cancer physicians in Australia are comfortable with shared treatment decision making (STDM), but their reported use of this approach is considerably less than their comfort with this concept.8 This article reports the perceived barriers and facilitators to treatment decision making identified by cancer physicians when consulting with newly referred or newly diagnosed cancer patients. We hypothesized that factors such as type of cancer treated; physician discipline, age, and experience; and patient caseload may influence experience of barriers to STDM. Interventions are proposed that may assist physicians to encourage patient involvement and reflection on treatment options.
Participant recruitment and procedures are reported in full elsewhere.8 Medical and radiation oncologists, hematologists, pediatric oncologists, and cancer surgeons practicing in Australia, managing people with five cancer sites (breast, colorectal, gynecologic, hematologic, or urologic), were invited to participate. Professional representative bodies were contacted to gain permission to invite group members to participate in this study. If this was granted and the addresses were provided, the research team mailed each physician a package including an invitation to participate in the survey intended to compare views of cancer physicians, an information sheet, a consent form, the questionnaire, and postage paid envelope. In the urologic cohort contact details were not provided, but the representative body sent the package to their members. The survey instrument was based on a structured questionnaire developed by Charles et al9,10 in Ontario, Canada, in 1998 through focus groups and pilot testing. The survey assessed the validity of the conceptual framework developed by Charles et al,11,12 both in its’ description of the meaning of STDM and in the agreement of practicing physicians to the key components. With permission, we used this questionnaire with some alterations. The questionnaire presented four examples of a treatment decision-making scenario, each one representing one of the following approaches: paternalistic, information sharing only, informed, and shared. The survey did not define shared decision making (SDM). This was a cross-sectional survey. Participants were asked to rate on a four-point Likert scale to what extent they experienced selected items as barriers or facilitators to decision making. Physicians treating breast or urologic cancer were also asked to report their support for a range of listed interventions to encourage patient involvement and reflect on treatment options during the decision-making process.
Data Analysis
Participants Of 1,062 eligible physicians approached, 624 completed surveys were returned, representing a response rate of 59%. This low response rate is comparable to the mean response rate of 54% for physician surveys.13 Twenty surveys completed by clinicians who reported they did not treat patients in the five targeted tumor groups were excluded from statistical analysis. Within the nonrespondents, tumor specialties were as follows: breast, 30%; colorectal, 10%; gynecologic, 2%; hematologic, 9%; urologic, 42%; and unknown, 7%. The high nonresponder rate in the urologic cohort may reflect the mailout method in this group. Excluding the urologic cohort on whom we had no sex information, 89% of the nonresponders were men. Comparison of this characteristic with the study sample reveals no notable differences. Table 1 shows the demographics of the 604 participating clinicians. Male clinicians (83.3%) made up the larger proportion of the sample. Mean age was 50 years, and mean time since medical qualification was 26 years.
Perceived Barriers to Treatment Decision Making Experience of items perceived as barriers to treatment decision making is presented in Table 2. Response categories were on a four-point Likert scale: never, sometimes, often, and almost always. Items experienced by the most physicians as difficulties almost always were as follows: the physician having insufficient information to make a decision at the first consultation (6.6%) and insufficient time to spend with the patient (5.5%). Items perceived by the most physicians as never a hindrance to treatment decision making were as follows: the doctor experiencing difficulty in framing the treatment options for the patient (48.3%) and the patient's family overriding the decision-making process (43.7%). Most common difficulties reported often or almost always were as follows: the physician having insufficient information to make a decision at the first consultation (28.9%), insufficient time to spend with the patient (28.4%), patient having other health problems (27.9%), patient having misconceptions about their disease or treatment (27.1%), patient being indecisive (24.2%), patient being too anxious (23.6%), and patient not understanding the information given (20.6%).
On the basis of conceptual analysis and the correlational findings, exploratory factor analyses were conducted. Two items, cultural differences and patient receiving conflicting recommendations, were removed because of statistical reasons. The items of insufficient information and patient comorbidity were removed because of their relatively low factor loadings and because they did not fit conceptually with the other items loaded onto the patient difficulties factor. Confirmatory factor analysis produced a four-factor model, conceptually defined as follows: patient difficulties, preference for a not recommended treatment, system difficulties, and agenda-setting patient. The patient difficulties factor included items that covered misunderstanding or lack of understanding, anxiety, denial, withdrawal, and indecision. The patient preference for a not recommended treatment factor included items that indicated the patient did not want to accept the physician-recommended treatment, through refusal or request for another treatment. System issues included lack of time, misunderstanding of the nature of the consultation and the decision-making process, and physician difficulties in framing options. The fourth factor included items that reflected an agenda-setting patient, either by bringing too much information or by wishing to be more involved than the doctor was comfortable with. The factor loadings for this analysis confirmed the four-factor exploratory analysis. The analysis gave a  2 of 199.99 on 84 degrees of freedom, with a Steiger Lind root mean square error of approximation of 0.05, a root mean residual of 0.04, and a goodness of fit index of 0.98. These values indicate that the model provides a good approximation to the data. Factor loadings are presented in Table 3.
Predictors of Barriers to Treatment Decision Making To further explore the reported barriers, the variables relating to each factor were summed and a score was computed for each of the four factors. A higher score represented greater reporting of the factor as a difficulty. Mean scores are presented in Table 3. Results of univariate analysis are presented in Table 4. Urologic cancer physicians had the highest mean score for the patient difficulties factor. Breast cancer physicians had the lowest mean score for the preference for a not recommended treatment factor. Hematologists and gynecologic physicians had the highest mean score for system difficulties. Surgeons reported the highest mean score for patient difficulties, and hematologists reported the highest mean score for system difficulties. Mean scores for each of the factors were comparable, regardless of physicians’ reported usual approach to decision making. Physicians whose initial medical training was completed in Asia, Southeast Asia, or the Middle East had higher mean scores for three of the four factors: patient difficulties, system difficulties, and agenda-setting patient. Physicians reporting less direct patient contact per week and fewer new patients per month showed a higher mean score for the patient difficulties factor.
Our data showed a high correlation between the variables of age and years of experience (P < .01) and physician discipline and type of cancer treated (P < .01). We included only years of experience and type of cancer treated in regression analyses. Bivariate correlation of cancer, years of experience, doctor sex, and community size showed significant correlation with system difficulties (Table 4). New patients seen per month, cancer type treated, and hours of practice per week showed significant correlation for the patient preference for a not recommended option. Predictors of greater reporting of system difficulties were identified using linear regression (Table 5). Three variables included had a significant association with a higher system limitation score. Variables included in the model were years of experience, cancer type treated, and community size. The final model (R2 = 0.035) showed that physicians with less experience (P = .000) reported greater frequency of system difficulties as barriers to decision making with patients.
Facilitators to Treatment Decision Making Six possible facilitators were listed and the frequencies with which physicians found these helpful obtained. As responses were skewed, the original four categories on a Likert scale were collapsed into a dichotomous score. Those reported by physicians as experienced as helpful never or sometimes were recoded as not helpful, and those reported as experienced often and almost always helpful were recoded as helpful. The majority of physicians (98.2%) reported that the patient trusting them was helpful and that the patient being accompanied in the consultation (91.5%), the patient wanting to participate (82.0%), and providing written information were helpful (80.3%; Table 2).
Interventions to Encourage Patient Involvement and Reflection
We examined the reported barriers and facilitators to treatment decision making of a large sample of practicing cancer physicians across Australia. We hypothesized that physician characteristics, including discipline and the type of cancer treated, may influence experience and perception of factors as barriers to SDM. Lack of time was reported as one of the more frequently experienced barriers to treatment decision making by Australian cancer physicians. Physicians with less experience reported system issues as difficult more often. This may be because physicians with greater experience are more patient-centered in their approach to consultations or have become skilled at overcoming system issues. A number of studies have reported that collaborative decision making does not increase consultation time and may save time in the long run, because thorough initial discussion enables subsequent consultations to be more succinct.14-16 Moreover, addressing patient concerns may shorten initial consultations. Patient misunderstanding was considered one of the key difficulties when reaching treatment decisions. Quirt et al17 explored the level of agreement between patients and their physicians about diagnosis, treatment aims, and risks and benefits. They reported that many lung cancer patients did not fully comprehend their situation and that this was underestimated by their physicians. If this discrepancy between the patient understanding and physician awareness of patient understanding is duplicated in Australia, this barrier may be higher than reported by our participants. Our data showed that physicians reported that patient anxiety and misconception of their disease or treatment were two of the most frequent barriers to sharing decision making. This is not surprising, as emotion has a negative effect on many requirements for collaboration, making it more difficult for patients to take in information and consider their options. In the review of Gravel et al18 of barriers to implementing SDM in clinical practice, lack of applicability of the process owing to patient characteristics, such as anxiety, and characteristics of the clinical situation, such as lack of any real choice, also featured strongly. Our data show cancer physicians’ perceptions of what assists them when making treatment decisions. Patients’ trust was regarded as the key facilitator. Patients also nominate trust in their physician as one of the most important factors when reaching a treatment decision.19 As patient involvement and the consumer role in healthcare grows,20 the importance of physicians gaining their patients’ trust may increase. Previous studies suggest that trust develops through excellent communication and by responding to patient preferences for information and involvement.21-23 However, difficulties in defining and measuring trust complicate research into its’ effect on doctor decision-making style.24 Patients’ being accompanied at the consultation was viewed as helpful. This may reflect recognition of the two-fold benefit a companion brings for the patient (as a second pair of ears and as emotional support). However, the presence of a third person as a facilitator to patient involvement was supported less, implying that physicians are more comfortable with patients bringing a significant other to the consultation than having a third person present, perhaps a nurse or patient advocate. Perceived differences in the roles that a patient's family or significant other might play compared with that of a nurse practitioner are unclear. Perceived patient willingness to participate in the decision-making process was viewed by the majority of physicians as a facilitator and reflects the notion that for a shared approach, both parties have to be willing to participate.11,12 Some written information given to patients is not balanced or complete,25,26 yet our findings imply that patients and physicians may benefit from patients receiving written evidence-based information to reflect on the treatments being recommended. We found little support for interventions designed to prepare patients to be involved in decision making, with less than one third of respondents embracing booklets explaining clinical decision making, patient roles in decision making, or use of question prompt lists. Audio recording of consultations was supported by only 16% of physicians. A review of the effectiveness of providing consultation audio recording or written summaries showed that the majority of patients found them to be useful and that recall and satisfaction with information received were improved. Anxiety or depression were not influenced by receipt of an audio recording or summary.27 Our findings suggest this review's recommendation that physicians should consider offering patients audio recordings seems not to have been adopted in Australia. Translating research evidence into practice remains a challenge for researchers keen to implement effective communication tools. An investigation of barriers to uptake of patient decision aids in clinical practice reported that physicians’ lack of awareness and of the skills to use the tools effectively inhibited implementation.28 Training programs targeted at health professionals are recommended to address the issue of lack of familiarity,11,12,18,28 yet our data reveal that Australian cancer physicians have minimal interest in access to training to meet patient preferences for SDM. The difficulties identified in our survey and the limited support for strategies and interventions echo the views of Entwistle and Watt,29 who state that the clinician's motivation to involve patients is influential and that success in facilitating involvement is inextricably linked to this. The difficulties we report emphasize system and patient attributes that prove challenging for physicians. Insight into the difficulties that may be physician driven and reasons why physicians may be less supportive of patient involvement should be sought in further research. Changing established practice and successfully implementing evaluated communication interventions remains a challenge. Perhaps greater implementation success will be realized by aiming strategies at patients or at the system.30 In conclusion, our survey sought to identify physician-perceived barriers and facilitators to the treatment decision-making process in cancer consultations and to explore support for interventions to increase patient involvement and reflection on treatment options presented to them. What hinders? System issues, particularly time, are a barrier to shared treatment decision making, particularly for less experienced physicians. Patients who refuse the recommended treatment or wish to pursue a treatment not considered optimal are also seen as a barrier to STDM. What helps? A patient trusting their doctor is important, although how to create and build trust is not clear. The value of the patient being accompanied during a consultation and being provided with quality written information is strongly endorsed by surveyed physicians. Successful implementation of interventions to encourage patient involvement in decision making is a challenge. Interventions that physicians support need to be developed, as well as increasing physician access to and awareness of interventions already available. This study presents self-reported survey data on cancer physicians’ perceived barriers and facilitators to treatment decision making with their patients; the data does not necessarily represent the importance physicians place on these barriers and facilitators. The 59% response rate presents some limitation to the generalizability of the data. There is potential for bias to have influenced the results, because responders may be more interested in treatment decision-making approaches used with patients than are nonresponders. Consequently, the reported experience of difficulties to SDM may be under-reported, and support for SDM may be over-representative of the situation in practice. The survey did not detail specific decisions contexts, and physicians were asked to consider their approach to decision making with their newly referred or newly diagnosed cancer patients.
The author(s) indicated no potential conflicts of interest.
Conception and design: Heather L. Shepherd, Martin H.N. Tattersall, Phyllis N. Butow Financial support: Martin H.N. Tattersall, Phyllis N. Butow Administrative support: Heather L. Shepherd Provision of study materials or patients: Heather L. Shepherd Collection and assembly of data: Heather L. Shepherd Data analysis and interpretation: Heather L. Shepherd, Martin H.N. Tattersall, Phyllis N. Butow Manuscript writing: Heather L. Shepherd, Martin H.N. Tattersall, Phyllis N. Butow Final approval of manuscript: Heather L. Shepherd, Martin H.N. Tattersall, Phyllis N. Butow
We thank Cathy Charles, PhD, Associate Professor (Department of Clinical Epidemiology and Biostatistics, McMaster University) for permitting us to use the survey instrument and for her time spent discussing this project. We also thank the 624 doctors who completed and returned the survey to us.
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Brown RF, Butow PN, Sharrock MA, et al: Education and role modelling for clinical decisions with female cancer patients. Health Expect 7:303-316, 2004[CrossRef][Medline] 2. Beaver K, Luker KA, Owens RG, et al: Treatment decision making in women newly diagnosed with breast cancer. Cancer Nurs 19:8-19, 1996[Medline] 3. Degner LF, Kristjanson LJ, Bowman D, et al: Information needs and decisional preferences in women with breast cancer. JAMA 277:1485-1492, 1997 4. Bruera E, Willey JS, Palmer JL, et al: Treatment decisions for breast carcinoma: Patient preferences and physician perceptions. Cancer 94:2076-2080, 2002[CrossRef][Medline] 5. Shotton L: Health care law and ethics. Katoomba, NSW, Australia, Social Science Press, 1997 6. Jahng KH, Martin LR, Golin CE, et al: Preferences for medical collaboration: Patient-physician congruence and patient outcomes. Patient Educ Couns 57:308-314, 2005[CrossRef][Medline] 7. Anderson R, Funnell M, Butler P, et al: Patient Empowerment: Results of a randomized controlled trial. Diabetes Care 18:943-949, 1995[Abstract] 8. Shepherd HL, Tattersall MHN, Butow PN: The context influences doctors’ support of shared decision making in cancer care. Br J Cancer 97:6-13, 2007[CrossRef][Medline] 9. Charles C, Gafni A, Whelan T: Self-reported use of shared decision-making among breast cancer specialists and perceived barriers and facilitators to implementing this approach. Health Expect 7:338-348, 2004[CrossRef][Medline] 10. Charles CA, Whelan T, Gafni A, et al: Shared treatment decision making: What does it mean to physicians? J Clin Oncol 21:932-936, 2003 11. 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 44:681-692, 1997[CrossRef][Medline] 12. Charles C, Gafni A, Whelan T: Decision-making in the physician-patient encounter: Revisiting the shared treatment decision-making model. Soc Sci Med 49:651-661, 1999[CrossRef][Medline] 13. Asch DA, Jedrziewski MK, Christakis NA: Response rates to mail surveys published in medical journals. J Clin Epidemiol 50:1129-1136, 1997[CrossRef][Medline] 14. Edwards A, Elwyn G, Mulley A: Explaining risks: Turning numerical data into meaningful pictures. BMJ 324:827-830, 2002 15. Say RE, Thomson R: The importance of patient preferences in treatment decisions: Challenges for doctors. BMJ 327:542-545, 2003 16. Greenfield S, Kaplan S, Ware JE Jr: Expanding patient involvement in care: Effects on patient outcomes. Ann Intern Med 102:520-528, 1985 17. Quirt CF, Mackillop WJ, Ginsburg AD, et al: Do doctors know when their patients don't? A survey of doctor-patient communication in lung cancer. Lung Cancer 18:1-20, 1997[CrossRef][Medline] 18. Gravel K, Legare F, Graham I: Barriers and facilitators to implementing shared decision-making in clinical practice: A systematic review of health professionals’ perceptions. Implement Sci 1:16, 2006[CrossRef][Medline] 19. Salkeld G, Solomon M, Short L, et al: A matter of trust: Patient's views on decision-making in colorectal cancer. Health Expect 7:104-114, 2004[CrossRef][Medline] 20. Coulter A: Paternalism or partnership? BMJ 319:719-720, 1999 21. Keating NL, Green DC, Kao AC, et al: How are patients’ specific ambulatory care experiences related to trust, satisfaction, and considering changing physicians? J Gen Intern Med 17:29-39, 2002[CrossRef][Medline] 22. Thom DH, Physicians STS: Physician behaviors that predict patient trust. J Fam Pract 50:323-328, 2001[Medline] 23. Trachtenberg F, Dugan E, Hall MA: How patients’ trust relates to their involvement in medical care: Trust in the medical profession is associated with greater willingness to seek care and follow recommendations. J Fam Pract 54:344-352, 2005[Medline] 24. Entwistle V: Trust and shared decision-making: An emerging research agenda. Health Expect 7:271-273, 2004[CrossRef][Medline] 25. Braddock III CH, Edwards KA, Hasenberg NM, et al: Informed decision making in outpatient practice: Time to get back to basics. JAMA 282:2313-2320, 1999 26. Coulter A: Patient information and shared decision-making in cancer care. Br J Cancer 89:S15-S16, 2003 27. Scott JT, Entwistle VA, Sowden AJ, et al: Giving tape recordings or written summaries of consultations to people with cancer: A systematic review. Health Expect 4:162-169, 2001[CrossRef][Medline] 28. O'Donnell S, Cranney A, Jacobsen MJ, et al: Understanding and overcoming the barriers of implementing patient decision aids in clinical practice. J Eval Clin Pract 12:174-181, 2006[CrossRef][Medline] 29. Entwistle VA, Watt IS: Patient involvement in treatment decision-making: The case for a broader conceptual framework. Patient Educ Couns 63:268-278, 2006[CrossRef][Medline] 30. Holmes-Rovner M, Valade D, Orlowski C, et al: Implementing shared decision-making in routine practice: Barriers and opportunities. Health Expect 3:182-191, 2000[CrossRef][Medline] Submitted July 16, 2007; accepted December 12, 2007.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
