Lung cancer symptom

Study objectives: To quantify clinician knowledge and bias regarding the role of chemotherapy for stage IV non-small cell lung cancer (NSCLC).

Design, setting, and participants: A 16-question, multiple-choice questionnaire was sent to all Australian general internists, pulmonary and palliative care physicians, medical and radiation oncologists, and thoracic surgeons to assess beliefs concerning the role of chemotherapy in metastatic NSCLC. An overall assessment of "pessimism" and "optimism" regarding the role of chemotherapy in metastatic NSCLC was made, and knowledge of specific outcome measures was evaluated.

Measurements and results: A total of 1,325 questionnaires were mailed, with 679 replies (51%) received and 544 replies (41%) assessable. Overall, 60% of respondents were deemed to have good knowledge. There was a wide variation in knowledge between specialist groups (p < 0.0001), with more medical oncologists (76%) but fewer thoracic surgeons (35%) and general internists (50%) with good knowledge. Fewer medical oncologists (6%) were classified as pessimistic compared with palliative care physicians (31%), radiation oncologists (28%), or pulmonary physicians (22%). Sixty-eight percent of respondents agreed that most patients receiving chemotherapy have symptomatic improvement. More medical oncologists (77%) and pulmonary physicians (73%), but fewer general internists (55%) and palliative care physicians (57%) agreed with this. Medical oncologists were far more likely to agree that chemotherapy was of benefit in patients aged [greater than or equal to] 70 years compared with any of the other specialist groups.

Conclusions: There were significant differences regarding the perceived role of chemotherapy in metastatic disease between the various specialty groups involved in the treatment of NSCLC. Many clinicians had a poor understanding of contemporary data regarding the use of chemotherapy in metastatic NSCLC. This study raises substantial issues regarding the beliefs of clinicians treating NSCLC and emphasizes the importance of multidisciplinary assessment.

Key words: attitudes; clinical practice patterns; non-small cell lung cancer; physician knowledge; prognosis; questionnaire

Abbreviations: BSC = best supportive care; ECOG = Eastern Cooperative Oncology Group; NSCLC = non-small cell lung cancer; PBC = platinum-based combination chemotherapy; PS = performance status; QOL = quality of life

**********

Lung cancer is the most common cause of cancer morality in Australia, with almost 8,000 new cases diagnosed in 1999, accounting for 19% of all cancer-related deaths. (1) The role of chemotherapy in metastatic (stage IV) non-small cell lung cancer (NSCLC) has evolved rapidly over the past decade. Older trials (2) using alkylating agents showed a detrimental effect on survival, contributing to a sense of pessimism regarding chemotherapy among clinicians treating NSCLC. However, the introduction of newer agents has resulted in level 1 evidence supporting a survival benefit from platinum-based combination chemotherapy (PBC) regimens compared with best supportive care (BSC). (2-4) Although the use of chemotherapy in this setting appears to be increasing, only a subset of potentially suitable patients ever receive chemotherapy. (5) This may be because of patient preference or unsuitability that may be due to severe illness or comorbidities, but it is possible that some patients who could benefit from chemotherapy may not receive it because of physician bias.

Clinicians in the United States have previously been surveyed to assess their perceptions and beliefs regarding treatment options in NSCLC. In 1998, Perez (6) reported significant variation between specialities regarding the use of chemotherapy for metastatic NSCLC, with the majority recommending only supportive care, except for medical oncologists who predominantly recommended chemotherapy. Schroen et al (7) surveyed pulmonologists and thoracic surgeons in 2000, and reported only one third of respondents believed chemotherapy conferred a survival benefit in stage IV NSCLC. Surveys (8,9) of clinicians in the mid-1990s in Italy and the United Kingdom showed a similar wide range of beliefs and a low use of chemotherapy for metastatic NSCLC.

No surveys have previously been reported specifically assessing clinician beliefs regarding the management of metastatic NSCLC. We hypothesized that there is a diverse range of beliefs between specialist groups, and thus sought to assess and quantify these beliefs.

MATERIALS AND METHODS

Participants and Questionnaire Tool

A 16-question, multiple-choice questionnaire (see "Appendix") was sent to all Australian consultant general internists (physicians), pulmonary and palliative care physicians, medical and radiation oncologists, and thoracic surgeons. Eligible clinicians were identified by the databases of the Internal Medicine Society of Australia and New Zealand, the Thoracic Society of Australia and New Zealand, the Australasian Chapter of Palliative Medicine of The Royal Australasian College of Physicians, the Medical Ontology Group of Australia, the Faculty of Radiation Oncology of the Royal Australian & New Zealand College of Radiologists, and the Royal Australasian College of Surgeons, respectively. Trainees and clinicians not in active practice were excluded.

The questionnaire included demographic data and an estimate of the number of patients with lung cancer seen annually. Questions to assess clinician knowledge of survival and response rates to PBC for metastatic NSCLC were asked, along with assessment of beliefs regarding the role of chemotherapy for symptom relief and in certain subgroups of patients.

The questions were designed by consensus opinion of a group including representatives from the disciplines of medical oncology, radiation ontology, pulmonary medicine, and thoracic surgery. The questions were specifically focused on chemotherapeutic management of stage IV NSCLC. Completed questionnaires were analyzed if they were returned within 6 weeks after mailing. Respondents were deemed assessable if they worked in clinical practice with adult patients and saw at least one patient a year with metastatic NSCLC.

Statistical Analysis

Data were analyzed comparing beliefs regarding chemotherapy and clinician characteristics. The results were analyzed using Pearson's [chi square] test with a two-sided analysis and deemed statistically significant if p < 0.05. To allow for multiple pairwise comparisons between specialist groups, comparisons were only considered statistically significant when using the procedure of Benjamini and Hochberg. (10) Statistical analysis was performed using computer software (Microsoft Excel 97: Microsoft; Redmond, WA).

An assessment of "knowledge" regarding survival and response rate was prospectively defined based on the summation of five answers, scoring 1 point for each correct answer. Consultants were deemed to have "good" knowledge if they scored [greater than or equal to] 3. Correct answers were determined from recently published, large-phase III trials and meta-analyses. (2-4,11-15) The correct answers, for a patient with newly diagnosed metastatic NSCLC who is otherwise well and of Eastern Cooperative Oncology Group (ECOG) performance status (PS) 0 to 2 are as follows: median survival without PBC, 3 to 6 months; 1-year survival without PBC, < 15%; median survival with PBC, 7 to 11 months; 1-year survival with PBC, 31 to 45%; and objective response rate to PBC, 15 to 30%.

The summation of seven answers was prospectively defined to produce an overall assessment of "pessimism" or "optimism" regarding the role of chemotherapy in metastatic NSCLC. Points were awarded for each answer from -2 for a very pessimistic response, through +2 for a very optimistic response (Table 1). Respondents were considered pessimistic if their total score was [less than or equal to] -3, and optimistic if [greater than or equal to] +3.

RESULTS

A total of 1,325 questionnaires were sent. Of the 679 replies received (51%), 544 respondents (41%) were assessable. Demographics are detailed in Table 2.