1 Department of Radiation Oncology, Brigham and Womens Hospital and Dana-Farber Cancer Institute, Boston, MA; 2 Department of Biostatistical Sciences, 3 Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
Received 29 April 2003; revised 27 August 2003; accepted 4 September 2003
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The aim of this study was to explore variation in practice patterns and identify factors associated with physicians treatment decisions for early-stage Hodgkins disease.
Methods:
We conducted a one-time mail survey of oncologists randomly selected from directories of national oncology societies (n = 207) and Hodgkins disease experts (n = 147). The survey included questions on (i) physician factors, (ii) preferred treatment choices for six case scenarios of early-stage Hodgkins disease that varied by patient factors, and (iii) thresholds for changing treatment recommendations.
Results:
The response rate was 50%. For non-bulky Hodgkins disease, 69% of respondents chose combined modality therapy (CMT). On multivariate analysis, physician factors that independently predicted for choice of CMT included a high Hodgkins disease case load (P = 0.02) and a high percentage of patients enrolled in clinical trials (P = 0.05). Radiation oncologists had a lower threshold for adding radiation therapy (P = 0.02). More experience with second malignancy cases and longer time in practice were associated with a higher threshold for adding radiation therapy (P = 0.04 and P = 0.008, respectively). In stratified analyses, treatment decisions of non-experts were significantly influenced by physician factors, but not by patient factors. Conversely, choices of Hodgkins disease experts were insensitive to all physician factors, but experts were significantly more likely to select chemotherapy alone in young women and CMT in older patients.
Conclusions:
Our results indicate that physician factors including practice type and experience may in part explain variation in practice pattern for Hodgkins disease therapy. Hodgkins disease experts are more likely to tailor therapy according to individual patient factors.
Key words: Hodgkins disease, physician survey, treatment decision
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
In the American College of Radiology Appropriateness Criteria Index, the criteria scores for CMT, radiation therapy alone and chemotherapy alone were 8, 46 and 2, respectively, for the treatment of early-stage Hodgkins disease (score of 9 being the most appropriate and score of 1 being the least appropriate) [10]. In the National Comprehensive Cancer Network Practice Guidelines (NCCN), based on strength of evidence and degree of consensus, CMT and radiation therapy alone were both listed as category 1 for early-stage Hodgkins disease, while chemotherapy alone was listed as category 3 (category 1, uniform NCCN consensus; category 3, major NCCN disagreement) [11]. Guidelines published by the United States National Cancer Institute in the management of stage III Hodgkins disease are illustrated in Table 2. Even in the presence of these guidelines, considerable heterogeneity exists in the treatment for early-stage Hodgkins disease. This may in part be because survival differences have never been demonstrated among the various treatment options. In addition, each treatment option is associated with a distinctive set of risks and benefits, and individual physicians may weigh different end points differently. In this study, we sought to determine the concordance of treatment recommendations by practicing oncologists with the existing guidelines for early-stage Hodgkins disease, and the variation in practice patterns among physicians. We also explored factors that influence a physicians thought process and bias in rendering a treatment recommendation for this patient population.
|
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The first part of the survey included questions on the physicians background. The second part of the survey solicited opinions about Hodgkins disease management through a series of clinical vignettes. The first vignette described a patient with early-stage Hodgkins disease, and asked the respondent to choose among treatment options of CMT, radiation therapy alone or chemotherapy alone. Six versions of the vignette were then presented, varying in gender, patient age and disease bulk. In the third part of the survey, physicians were asked to identify the threshold values at which their recommendations would change. Specifically, they were asked how large a reduction in the relapse rate they would require before adding chemotherapy to radiation, or adding radiation therapy to chemotherapy. They were also asked how large a reduction in the second-malignancy risk they would require to omit radiation therapy and treat with chemotherapy alone, assuming radiation exposure is the main cause of second malignancy after Hodgkins disease. Finally, they were asked how much improvement in overall survival they would require to recommend CMT over chemotherapy alone. The survey was piloted in a sample of 10 oncologists at the Dana-Farber Cancer Institute and Beth Israel Deaconess Medical Center and was modified to reduce ambiguity based on their feedback.
Potential respondents were first contacted by E-mail. The E-mail provided a brief description of the study, asked respondents several screening questions, and offered them the opportunity to opt out of the study. Physicians were excluded if they reported that they were in training, spent less than 5 h per week in clinical practice or had managed fewer than five Hodgkins disease cases in the last 3 years. The survey, along with an explanatory letter, was mailed to eligible physicians who either did not opt out or who actively agreed to participate. Non-responders were sent a reminder E-mail 23 months after the first mailing. No incentives were offered to participating physicians.
To explore the association between factors included in the survey and treatment recommendation made by physicians, univariate analysis using Fishers exact test and generalized logits models were used. To evaluate the consistency of physicians treatment recommendations, kappa statistics were used to test the agreement of treatment recommendations according to the age, gender and disease burden of patients described in the vignettes. The larger the percentage of consistency or the smaller the P value, the higher is the extent of agreement with respect to the factor under consideration. P 0.05 was considered statistically significant. All the tests were two-sided.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The median number of years in practice for all physicians was 19 (range 243). The median number of Hodgkins disease cases seen each year was 12 (range 2250). The median number of second-malignancy deaths under the care of the physicians was two (range 060). Among the medical and pediatric oncologists, 75% had a radiation oncology facility within their institution. The remainder of the baseline characteristics of the physicians are summarized in Table 3.
|
|
|
Hodgkins disease experts were more likely to recommend CMT and less likely to choose either radiation therapy alone or chemotherapy alone (Table 6, section A). Medical oncologists were more likely to recommend CMT or chemotherapy alone. In contrast, radiation therapy alone was favored by more radiation oncologists than medical oncologists (Table 6, section B). Compared with physicians from the United States, non-USA physicians were more likely to recommend CMT and less likely to choose radiation therapy alone (Table 6, section C). Academic physicians were also more likely to choose CMT over radiation therapy or chemotherapy alone (Table 6, section D). Other physician characteristics that were associated with a significantly higher likelihood of recommending CMT included a higher Hodgkins disease case load (defined as seeing more than 12 cases of Hodgkins disease each year), practice in a multidisciplinary setting, having a high percentage of patients enrolled in clinical trials, and familiarity with Hodgkins disease clinical guidelines. Physicians with more experience in handling second malignancy cases (defined as having taken care of more than two patients who died of a second malignancy after Hodgkins disease) were significantly less likely to recommend radiation therapy alone and more likely to recommend chemotherapy alone or CMT. Length of time in practice and having a radiation oncology facility in the vicinity did not significantly influence treatment recommendations.
|
We also explored the association between physician-related factors and thresholds for adding radiation therapy or chemotherapy. Compared with medical oncologists, radiation oncologists had a lower threshold for adding radiation therapy to chemotherapy in the scenario based on improvement in overall survival (P = 0.02). Physicians with more experience in dealing with second malignancy cases had a higher threshold for adding radiation therapy to chemotherapy alone based on improvement in relapse risk (P = 0.04). Physicians who had been in practice longer (19 years or more) had a higher threshold for adding radiation therapy based on relapse rate (P = 0.008) and overall survival (P = 0.01) and a higher threshold for adding chemotherapy to radiation therapy alone (P = 0.04). Thresholds for omitting radiation therapy based on improvement in 25-year-old second-malignancy risk were not sensitive to any of the factors considered. Area of expertise, Hodgkins disease case load, practice setting of academic versus non-academic, and country of practice did not significantly influence thresholds for changing treatment recommendations in all scenarios.
Patient-related factors associated with treatment recommendations
We examined the impact of patient-related factors, including age and gender, on the treatment choices. Among all physicians, treatment recommendation did not differ by patient age. Chemotherapy alone was recommended more often for young female than for young male patients (15% versus 10%), and CMT was recommended less often (65% versus 72%).
Stratified analyses of Hodgkins disease experts versus non-experts
Physicians who were identified as experts in Hodgkins disease differed significantly from non-experts in their demographic characteristics, as summarized in Table 7. Because of their distinctive baseline characteristics, we separately explored the influence of physician- and patient-related variables on the treatment recommendations and thresholds of the Hodgkins disease experts and non-experts. Among Hodgkins disease experts, in whom less variability in treatment recommendations was observed, none of the demographic variables was found to significantly influence their treatment choices or thresholds for changing treatment recommendations. In contrast, a number of physician factors, including country of practice, percentage of patients enrolled in clinical trials, length of time in practice, and experience with second malignancy cases significantly influenced the treatment choices of non-experts.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Many of the physician characteristics explored in this survey were interrelated. After adjustment for other factors, a high Hodgkins disease patient load was the single most important independent predictor of consistent recommendation of CMT over chemotherapy alone or radiation therapy alone for both male and female patients and both younger and older patients. This finding indicates that less-experienced physicians may still be influenced by the more traditional approach of radiation therapy alone in early-stage Hodgkins disease, and may not yet have embraced CMT as the standard of care. A higher percentage of patients enrolled in clinical trials was also independently associated with a preference for CMT in the young male and female patients. It is possible that these physicians may have been influenced to view CMT as the standard of care because it is a component of most recent and ongoing trials on early-stage Hodgkins disease.
One unique aspect of this survey study was that we explored how thresholds for changing treatment recommendations varied according to physician characteristics and the end points under consideration. Medical oncologists were more ready to add chemotherapy or omit radiation therapy, while radiation oncologists were more ready to add radiation therapy and less ready to omit radiation therapy. Such specialty-based biases have also been observed in other disease sites. In the management of genitourinary malignancies, studies have found that urologists tend to favor surgery, while radiation oncologists tend to choose radiation therapy over surgery [12, 13]. In an international survey on treatment for gastric lymphoma, hematologists and medical oncologists were found to prefer conservative treatment, while gastroenterologists were more inclined to choose a surgical approach [14]. In the current study, longer time in practice was associated with significantly more stringent requirements for adding radiation therapy or chemotherapy. Presumably, physicians who have been practicing longer have had to deal with more long-term effects of cancer treatment; they would therefore tend to demand a greater gain before they are willing to add another modality to the overall treatment. This is also supported by our finding that physicians who have seen a greater number of fatal second malignancies after Hodgkins disease have a significantly higher threshold for adding radiation therapy to chemotherapy based on relapse risk. It is of interest that when physicians were asked of their thresholds for adding or omitting a treatment modality, none of the physician characteristics seemed to be sensitive to the long-term risk of second malignancy. This finding indicates that when physicians render a treatment recommendation, the more immediate and directly disease-related end points may be foremost on their mind, while the delayed end points have less of an impact on their decision.
It is reassuring that none of the physicians recommended radiation therapy alone for patients with bulky disease, as radiation therapy alone has been shown to be associated with a high risk of relapse in this patient population [1517]. There is a trend for more frequent recommendation of chemotherapy alone in young female patients, although the difference did not reach statistical significance. Treatment recommendations also did not differ significantly according to patient age.
Further analysis stratifying physicians into Hodgkins disease experts versus non-experts showed that the treatment choices and thresholds to change treatment recommendation of the experts were less prone to influence by their demographic characteristics than the non-experts. However, the experts were more inclined to tailor their recommendations according to patient age and gender. Among the Hodgkins disease experts, CMT was favored in older patients. A number of studies have shown that older patients with Hodgkins disease tend to fare worse for a variety of reasons [1822]. CMT offers these patients higher upfront disease control, and also allows the possibility of dose reductions when both modalities are used. The Hodgkins disease experts were also more likely to recommend chemotherapy alone in young female patients. This finding is consistent with the increasing data on the significantly elevated risk of breast cancer in women who received radiation treatment for Hodgkins disease at a young age [2328].
In this survey, only three main treatment approaches were offered as choices. More detailed treatment options such as types and number of cycles of chemotherapy, radiation field size and radiation dose were not available. Also, patients were stratified by factors including age, gender and disease bulk. However, other important information which may affect treatment recommendations such as performance status, presence or absence of constitutional symptoms, extranodal disease and sedimentation rates were not provided. In the design of the survey, we had to weigh the level of detail in the case scenarios against the length of the survey, which can negatively affect the response rate, and as such, only key variables were included.
It is encouraging that the degree of acceptance of the three treatment options coincides well with current existing guidelines for early-stage Hodgkins disease. However, we did find that physician choices and thresholds for adding or omitting part of the treatment were significantly influenced by experience, practice type, setting and environment and the immediacy of the end point under consideration. Understanding how physicians arrive at a treatment recommendation, especially in the absence of a clear benefit of one option over another, is important. Patients rely heavily on the opinions of their physicians. In one study, using hypothetical scenarios, it was shown that even when one management option was obviously superior to another, patients choices were significantly influenced by the recommendation of the physician, even if the recommendation clearly did not maximize their health [29]. Physicians may benefit from being made aware of external factors that affect their recommendations to patients, in order to promote uniform quality of care. In this study, Hodgkins disease patient load was the single most important independent predictor for choice of CMT. Furthermore, Hodgkins disease experts tended to choose treatment tailored to patient-related factors and the choices were relatively insensitive to their personal background. Together, these findings indicate that patients may be more likely to receive objective and consistent treatment recommendations that are also closest to current existing guidelines if they receive care from physicians with more extensive experience with Hodgkins disease and its management.
![]() |
FOOTNOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Advani RH, Horning SJ. Treatment of early-stage Hodgkins disease. Semin Hematol 1999; 36: 270281.[ISI][Medline]
3. Ng AK, Mauch PM. Controversies in early-stage Hodgkins disease. Oncology 2002; 16: 588595, 598 [Discussion 600, 605, 609618].
4. Hoppe RT, Coleman CN, Cox RS et al. The management of stage III Hodgkins disease with irradiation alone or combined modality therapy: the Stanford experience. Blood 1982; 59: 455465.[Abstract]
5. Noordijk EM, Carde P, Mandard AM et al. Preliminary results of the EORTC-GPMC controlled clinical trial H7 in early-stage Hodgkins disease. EORTC Lymphoma Cooperative Group. Groupe Pierre-et-Marie-Curie. Ann Oncol 1994; 5: 107112.[Medline]
6. Specht L, Gray RG, Clarke MJ, Peto R. Influence of more extensive radiotherapy and adjuvant chemotherapy on long-term outcome of early-stage Hodgkins disease: a meta-analysis of 23 randomized trials involving 3888 patients. International Hodgkins Disease Collaborative Group. J Clin Oncol 1998; 16: 830843.[Abstract]
7. Press OW, LeBlanc M, Lichter AS et al. Phase III randomized intergroup trial of subtotal lymphoid irradiation versus doxorubicin, vinblastine, and subtotal lymphoid irradiation for stage IA to IIA Hodgkins disease. J Clin Oncol 2001; 19: 42384244.
8. Straus D, Yahalom J, Zelenetz A et al. Results of a prospective randomized trial of doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) alone vs ABVD + radiation therapy for early stage non bulky Hodgkins disease. The American Society of Hematology 43th Annual Meeting, Orlando, FL, USA 2001; 769a (Abstr 3201).
9. Nachman JB, Sposto R, Herzog P et al. Randomized comparison of low-dose involved-field radiotherapy and no radiotherapy for children with Hodgkins disease who achieve a complete response to chemotherapy. J Clin Oncol 2002; 20: 37653771.
10. Mauch PM, Connors JM, Dosoretz DE et al. American College of Radiology ACR Appropriateness Criteria. Favorable prognosis stage I and II Hodgkins disease treatment guidelines. http://www.acr.org 1998.
11. Hoppe RT, Abrams RA, Bernstein SH et al. National Comprehensive Cancer Network. Hodgkins Disease. Version 1. http://www.nccn.org/physician_gls/index.html 2002.
12. Moore MJ, OSullivan B, Tannock IF. How expert physicians would wish to be treated if they had genitourinary cancer. J Clin Oncol 1988; 6: 17361745.[Abstract]
13. Fowler FJ Jr, McNaughton Collins M, Albertsen PC et al. Comparison of recommendations by urologists and radiation oncologists for treatment of clinically localized prostate cancer. J Am Med Assoc 2000; 283: 32173222.
14. de Jong D, Aleman BM, Taal BG, Boot H. Controversies and consensus in the diagnosis, work-up and treatment of gastric lymphoma: an international survey. Ann Oncol 1999; 10: 275280.[Abstract]
15. Liew KH, Easton D, Horwich A et al. Bulky mediastinal Hodgkins disease management and prognosis. Hematol Oncol 1984; 2: 4559.[Medline]
16. Hughes-Davies L, Tarbell NJ, Coleman CN et al. Stage IAIIB Hodgkins disease: management and outcome of extensive thoracic involvement. Int J Radiat Oncol Biol Phys 1997; 39: 361369.[ISI][Medline]
17. Aviles A, Delgado S. A prospective clinical trial comparing chemotherapy, radiotherapy and combined therapy in the treatment of early stage Hodgkins disease with bulky disease. Clin Lab Haematol 1998; 20: 9599.[CrossRef][ISI][Medline]
18. Austin-Seymour MM, Hoppe RT, Cox RS et al. Hodgkins disease in patients over sixty years old. Ann Intern Med 1984; 100: 1318.[ISI][Medline]
19. Rossi Ferrini P, Bosi A, Casini C et al. Hodgkins disease in the elderly: a retrospective clinicopathologic study of 61 patients aged over 60 years. Acta Haematol 1987; 78: 163170.[CrossRef][ISI][Medline]
20. Bosi A, Ponticelli P, Casini C et al. Clinical data and therapeutic approach in elderly patients with Hodgkins disease. Haematologica 1989; 74: 463473.[ISI]
21. Diaz-Pavon JR, Cabanillas F, Majlis A, Hagemeister FB. Outcome of Hodgkins disease in elderly patients. Hematol Oncol 1995; 13: 1927.[ISI][Medline]
22. Weekes CD, Vose JM, Lynch JC et al. Hodgkins disease in the elderly: improved treatment outcome with a doxorubicin-containing regimen. J Clin Oncol 2002; 20: 10871093.
23. Yahalom J, Petrek JA, Biddinger PW et al. Breast cancer in patients irradiated for Hodgkins disease: a clinical and pathologic analysis of 45 events in 37 patients. J Clin Oncol 1992; 10: 16741681.[Abstract]
24. Hancock SL, Tucker MA, Hoppe RT. Breast cancer after treatment of Hodgkins disease. J Natl Cancer Inst 1993; 85: 2531.[Abstract]
25. Bhatia S, Robison LL, Oberlin O et al. Breast cancer and other second neoplasms after childhood Hodgkins disease. N Engl J Med 1996; 334: 745751.
26. Aisenberg AC, Finkelstein DM, Doppke KP et al. High risk of breast carcinoma after irradiation of young women with Hodgkins disease. Cancer 1997; 79: 12031210.[CrossRef][ISI][Medline]
27. Wolden SL, Hancock SL, Carlson RW et al. Management of breast cancer after Hodgkins disease. J Clin Oncol 2000; 18: 765772.
28. Cutuli B, Borel C, Dhermain F et al. Breast cancer occurred after treatment for Hodgkins disease: analysis of 133 cases. Radiother Oncol 2001; 59: 247255.[CrossRef][ISI][Medline]
29. Gurmankin AD, Baron J, Hershey JC, Ubel PA. The role of physicians recommendations in medical treatment decisions. Med Decis Making 2002; 22: 262271.