 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Medicare Drug Reimbursements: A Broken System for Patients and Taxpayers
Subcommittee on Oversight and Investigations
Dr. Ezekiel Emanuel
There is substantial concern
about end-of-life care provided to Americans. In particular, a number of
commentators are concerned that dying cancer patients are frequently
overtreated with chemotherapy. Critics contend that many oncologists
overtreat dying patients with chemotherapy because they are reluctant to
accept death and apprehensive about discussing end-of-life care. [i],[ii],[iii]
Indeed, some critics contend that oncologists prey on their patients'
vulnerability, implying that chemotherapy is the vehicle of hope, and pressing
them to try it before reconciling themselves to death.[iv]
Oncologists respond that it is terminally ill patients who demand treatment.
More importantly, oncologists contend that they use chemotherapy prudently in
patients at the end of life, when it is likely to provide symptom relief and
enhance dying patients' quality-of-life.
How can we determine if chemotherapy is used too frequently for terminally ill
cancer patients? There are no standards for the appropriate use of
chemotherapy at the end of life based upon either randomized controlled trials
or expert, consensus guidelines. While there are some data on treatment
of patients with metastatic cancers[v],
even basic data on how frequently cancer patients are given chemotherapy in
the months before death are lacking. To explore whether chemotherapy is used
prudently and rationally at the end of life, we separately examined its use
among Massachusetts and California Medicare beneficiaries who died of cancer
in 1996. Dividing patients into two groups according to whether they
died of cancers responsive or unresponsive to chemotherapy, we evaluated the
use of chemotherapy, and the expenditures in the last year of life. METHODS
Identifying
Cancer Decedents: To focus only on persons who died from cancer-not merely
with cancer-based on the primary cause of death listed in the death
certificate, we followed a 3-step process. First, in both Massachusetts
and California we studied fully entitled Medicare beneficiaries who died in
1996, were at least 66 years old at death and were not enrolled in
Medicare's End Stage Renal Disease program. Decedents 66 years of age were
selected to ensure we obtained a full year of Medicare expenditure data prior
to death. We studied all such decedents in Massachusetts and 5% in
California. Second, we merged HCFA's denominator files with each state's
1996 death certificate files. In Massachusetts, 42,452 Medicare
decedents met the criteria. In merging the files we used social security
number (SSN), date of birth (DOB), date of death (DOD) and sex. A match
was accepted if either of the following conditions was met: 1) there was a
perfect match on SSN and either sex or both DOB and DOD or 2) a match on at
least 7 of SSN digits and a perfect match on sex, DOB, and DOD. Of the
42,452 decedents, there was a match between the HCFA files and death
certificates for 39,447 (92.9%). Only beneficiaries continuously
enrolled in both Parts A and B Medicare insurance and who were not enrolled in
an managed care organization over the entire last 12 months of life were
retained, yielding 34,131 Massachusetts decedents. Third, we selected
the 7,919 decedents whose primary cause of death listed on the death
certificate was cancer. In
California, the same general protocol was applied to a random 5% of Medicare
enrollees yielding 4,715 total decedents overall, of which 956 died of cancer. Classifying Cancer Types: We
classified breast, colon, and ovarian cancers as chemotherapy responsive solid
cancers. Multiple chemotherapeutic agents shrink these cancers, and
randomized trials have shown chemotherapy to be effective in prolonging lives
of patients at least as adjuvant therapy. We classified pancreatic,
renal cell, hepatocellular, gallbladder, cancers, and melanoma as chemotherapy
unresponsive solid cancers. In 1996, these cancers were known to be
"refractory to virtually all chemotherapeutic agents" such that the
general consensus in standard textbooks is that "there are no particularly
compelling cytotoxic chemotherapeutic agents [with which] to treat" them.[vi] We
examined data for other cancers that we did not categorize as responsive or
unresponsive. For example, while prostate cancer is generally considered
a chemotherapy unresponsive solid cancer, hormonal injections may appear in
claims data as chemotherapy. To avoid uncertainty, prostate cancer is reported
separately. Lung cancer also examined separately because using claims data, it
is impossible to differentiate lung cancers into small cell and non-small cell
(NSCLC) tumors. Furthermore, while small cell cancers are chemotherapy
responsive, using chemotherapy for metastatic non-small cell lung cancers is
highly controversial.[vii]
Data suggest that chemotherapy for NSCLC extends life by 6 weeks and may
improve quality-of-life despite toxicities.[viii],[ix],[x]
Finally, hematological malignancies, encompassing both acute and chronic
leukemias, Hodgkin's disease, and all non-Hodgkin's lymphomas, were
analyzed separately. Although they are chemotherapy responsive, patients may
die acutely from treatment related toxicities. Identifying the Use of
Chemotherapy: Patients who had claims in the inpatient, outpatient or
physician/supplier Medicare files for chemotherapy agents, chemotherapy
administration, or the medical supervision of chemotherapy were considered to
as having received chemotherapy. The codes used were: intravenous
chemotherapy agents-HCPCS codes 964XX, 965XX, J9000-9999; chemotherapy
administration-IC Procedure 99.25, HCPCS codes Q0083-Q0085; medical
evaluation for chemotherapy-ICD Diagnosis V58.1, V66.2, and V67.2. It
is possible that our method of identifying chemotherapy missed patients who
received oral chemotherapeutic agents. Patients without claims using
these codes were classified as not having chemotherapy.
We examined chemotherapy use for decedents for 30- day periods from the date
of death back for 12 months. Expenditure Data: Total
expenditure is calculated as the sum of HCFA payments and payments from other
sources of insurance for Medicare covered services. The average payment
per person from other insurance accounts for only 0.15% of costs.
Expenditures for each decedent are calculated from 5 HCFA files: 1) Medicare
Provider Analysis and Review (MedPAR), including acute hospitalizations, long
term hospitalizations, and skilled nursing home care; 2) Hospital outpatient;
3) Part B physician-supplier; 4) Home health care; and 5) Hospice.
Durable medical equipment (DME) expenses were excluded, but in Massachusetts,
they contributed just $400 per person over the last year of life. RESULTS
Frequency
of Chemotherapy in the Last Months of Life: Figure 1 shows that 41% of our
study population in Massachusetts received chemotherapy in the last year of
life. Fully 33% of Massachusetts cancer decedents received chemotherapy
in the last 6 months of life, 23% in the last 3 months of life, and 9% of
cancer decedents received chemotherapy in the very last month of life.
Table 1 provides data on the proportion of terminally ill cancer patients
treated in Massachusetts with chemotherapy in the last 6, 3 and 1 months of
life. Patients who died of hematological malignancies received
chemotherapy most frequently, with more than half getting chemotherapy in the
last 6 months of life and 19% in the last month of life. Massachusetts
patients with chemotherapy unresponsive solid cancers received chemotherapy at
about the same frequency as patients with chemotherapy responsive solid
cancers (Table 1). Among patients with chemotherapy unresponsive solid cancers
taken together (pancreatic, hepatocellular, gallbladder, and renal cell
cancers and melanoma) 23% received chemotherapy in the last 3 months of life,
which was the same as the percentage of patients with chemotherapy responsive
cancers (breast, colon, ovarian) that received chemotherapy. An
interesting example of the use of chemotherapy at the end of life is
pancreatic cancer. In the last 6 months of life, 33% of Massachusetts patients
dying of pancreatic cancer received chemotherapy, 25% in the last 3 months,
and 8% in the last month of life. On May 15, 1996, the FDA approved
gemcitabine as the first agent shown to be effective in pancreatic cancer.
Prior to this date, when there were no effective agents, 28% of patients dying
of pancreatic cancer received chemotherapy in the last 6 months of life. After
May 15th, 37% received chemotherapy (one-sided p=0.04). A
comparison of the chemotherapy unresponsive melanoma and renal cell cancer
with chemotherapy responsive breast and colon cancers is also instructive. Of
patients dying of melanoma, 21% received chemotherapy in the last 3 months of
life and 10% in the last month of life. Similarly, among patients dying of
renal cell cancer, 22% received chemotherapy in the last 3 months of life and
7% in the last month of life. Surprisingly the frequency of chemotherapy for
dying breast and colon cancer patients was almost identical. 22% of patients
dying of breast cancer received chemotherapy in the last 3 months and 8% in
the last month of life. Similarly, 23% of patients dying of colon cancer
received chemotherapy in the last 3 months and 7% in the last month of life. There
are no substantial differences in the use of chemotherapy by sex (Table 1).
However, the use of chemotherapy at the end of life is age related. Among
Massachusetts patients 65-74 32% received chemotherapy in the last 3 months of
life, compared to 22% for patients 75 to 84 year old, and 11% for patients
over 85 years of age (Table 1). These variations by age were similar in
chemotherapy unresponsive and responsive solid cancers (Table 2). Overall, 13%
of 85 year olds with chemotherapy unresponsive solid cancers received
chemotherapy in the last 3 months of life compared to 10% of 85 year olds with
chemotherapy responsive solid cancers (Table 2). Number
of Months of Chemotherapy in the Last Months of Life: Among Massachusetts
patients who received chemotherapy in the last 6 months of life, 41% had a
short "trial," just one month or less of chemotherapy, with 36% receiving
chemotherapy for 1 to 3 months, 23% 4 or more months of chemotherapy (Table
3). The number of months of chemotherapy did not depend on sex, but did depend
upon age (Table 3). Importantly,
the chemotherapy responsiveness of the solid cancers was associated with a
difference in the number of months of chemotherapy provided to decedents
(Table 3). Among Massachusetts patients dying of chemotherapy unresponsive
tumors who received chemotherapy, over half received 1 month or less of
chemotherapy and 31% received chemotherapy for 1 to 3 months. Conversely,
among patients dying of chemotherapy responsive cancers who received
chemotherapy a third received 1 month or less of chemotherapy and 40% received
chemotherapy for 1 to 3 months of the last 6 months of life. Notably, 17% of
patients dying from chemotherapy unresponsive cancers had 4 or more months of
chemotherapy (Table 3). Returning
to patients with pancreatic cancer, 49% received chemotherapy for 1 month or
less, 34% for 1 to 3 months and 3% during each of the last 6 months. For
patients dying of breast cancer, 32% received chemotherapy for 1 month or
less, 39% for 1 to 3 months and 5% across all 6 final months. The
Use of Chemotherapy and Expenditures: Annual expenditures for dying
Massachusetts cancer patients who received chemotherapy in the last 6 months
of life were 32.5% higher than patients who did not receive chemotherapy
($39,707 v. $29,974) (Table 4). Annual expenditure for patients with
chemotherapy unresponsive cancers who received chemotherapy was $33,365 about
10% less than the expenditure for patients with chemotherapy responsive
cancers who received chemotherapy ($36,684). Expenditures for patients
with chemotherapy unresponsive cancers who received chemotherapy were 20% more
than for patients with the same cancers who did not receive chemotherapy
($33,365 v. $27,737), while expenditures for patients with chemotherapy
responsive cancers who received chemotherapy were 23.9% more than for patients
with the same cancers who did not receive chemotherapy ($36,684 v. $29,610). Comparison
with Cancer Decedents from California: We used decedents our sample of
956 cancer decedents from California to test whether our findings in
Massachusetts might apply more generally (Table 5). Among California
cancer decedents, 26% received chemotherapy in the last 6 months of life, 20%
in the last 3 months and 9% in the last month of life. Among decedents with
chemotherapy responsive tumors, 17% received chemotherapy in the last 3 months
of life compared to 20% for the chemotherapy unresponsive tumors. Similarly,
use of chemotherapy at the end of life was age related in California for both
chemotherapy responsive and unresponsive cancers. Among decedents aged
65-74, 26% of those with chemotherapy responsive tumors compared to 32% of
those with chemotherapy unresponsive tumors received chemotherapy in the last
3 months of life. Similarly, among decedents aged 75-84 19% of those
with responsive tumors compared to 18% of decedents with unresponsive tumors
received chemotherapy in the last 3 months of life. Overall, 25% of patients
with chemotherapy responsive tumors receiving chemotherapy received less than
1 month of chemotherapy while 35% of those with chemotherapy unresponsive
tumors did so. DISCUSSION
This
study provides insight into the frequency of use of chemotherapy at the end of
life. Overall 33% of Medicare patients dying of cancer in Massachusetts in
1996 received chemotherapy in the last 6 months of life and nearly a quarter
in the last 3 months. Most surprisingly, patients dying of chemotherapy
unresponsive cancers, such as pancreatic, gallbladder, renal cell, and
hepatocellular cancers, were just as likely to receive chemotherapy at the end
of life as patients dying of chemotherapy responsive cancers, such as breast,
colon, and ovarian cancers. This suggests overuse of chemotherapy at the
end of life, at least among patients with chemotherapy unresponsive cancers. Traditionally,
to document over- and underuse of health care services, studies compare claims
data with optimal practices established by randomized controlled trials or by
expert, consensus panels. Lacking randomized trials or consensus panels
to establish standards for the appropriate use of chemotherapy at the end of
life, we examined tumor responsiveness to chemotherapy. Cancers are
traditionally divided in those that are chemotherapy responsive, in which
chemotherapy can commonly induce complete and partial responses, compared to
those in which chemotherapy rarely leads to tumor shrinkage. In our
data, lack of responsiveness of the cancer to chemotherapy did not reduce the
prevalence of chemotherapy use. Patients with unresponsive cancers were
just as likely to receive chemotherapy in the last few months of life as
patients with chemotherapy responsive cancers. Indeed, patients with
unresponsive cancers were slightly more likely to receive chemotherapy than
patients with lung cancer in which data suggests chemotherapy in the last 6
months of life, may extend life by a few weeks and even palliate symptoms.
Although
patients dying of chemotherapy unresponsive solid cancers received
chemotherapy as frequently as those with responsive cancers, they received
fewer months of chemotherapy. This suggests some selectivity in the use of
chemotherapy at the end of life. It is possible that after one cycle of
therapy many patients and oncologists are convinced by ineffectiveness and/or
the side effects to stop treatment for chemotherapy unresponsive cancers.
Nevertheless, 17% of patients receiving chemotherapy for chemotherapy
unresponsive cancers received chemotherapy during four or more of the final 6
months of life. Many
reasons may explain the use of chemotherapy at the end of life for patients
with unresponsive cancers. The most reasonable explanation may be that
patients and families demand to at least "try" to see if chemotherapy
might shrink the cancer. Oncologists frequently meet patients for the first
time right after they have been newly diagnosed with chemotherapy unresponsive
tumors that present a bleak prognosis. These patients and their families
often want to try anything that might shrink their cancers. Indeed, data
suggest that cancer patients are willing to endure significant side effects
for very small prolongations in life.[xi],[xii]
Lacking an established relationship with the patient or family and confronting
an emotional demand to try anything, oncologists may acquiesce. One
cycle of chemotherapy is often sufficient for patients and families to adjust
and absorb the realities of the diagnosis, prognosis, and to realize the
ineffectiveness of the chemotherapy and the undesirable side effects.
That over half of the patients receiving chemotherapy for unresponsive cancers
received 1 month or less of chemotherapy strongly supports this explanation.
Obviously, additional research is necessary to provide insights into how much
of a role patient and family demand plays in the use of chemotherapy at the
end of life. Other
potential reasons for the use of chemotherapy at the end of life include
uncertain prognosis and time of death, uncertain responsiveness of the cancer
to chemotherapy, and use of experimental chemotherapies. These reasons
are unlikely to account for our data on chemotherapy unresponsive solid
cancers. While the exact date of death cannot be known in advance,
cancers, especially chemotherapy unresponsive solid cancers, are unlike the
terminal phases of COPD or heart failure; they tend to have a monotonic,
unremitting decline to death despite all interventions.[xiii]
Typically within the last three months of life, oncologists can predict, with
reasonable certainty that the patient will die in a few months regardless of
treatment. Furthermore, there is no real uncertainty about the chemotherapy
unresponsiveness of the solid tumors we classified as "unresponsive."
Finally, although some patients may be receiving experimental chemotherapy,
this is likely to be rare among Medicare beneficiaries who are often
ineligible due to age and comorbidities. Yet
another potential explanation for the use of chemotherapy for patients with
unresponsive cancers is that chemotherapy may improve quality of life and
palliate symptoms for dying patients even if it fails to prolong life or
shrink tumors.[xiv],[xv]
There are some data supporting the palliative effect of chemotherapy for lung
and colon cancer and some suggestions that this might also operate in ovarian
cancer.[xvi],[xvii],[xviii],[xix]
Frequently, emotional functioning and fatigue are the quality-of-life
subscales with the most improvement. That these improvements occur
without objective tumor responses suggests that they may be related to patient
expectations or possibly the placebo effect of chemotherapy, rather than any
biological impact.[xx]
The mechanism by which chemotherapy in terminal phases may palliate without
objectively shrinking cancers requires further research. The
similar frequency of chemotherapy use regardless of the responsiveness of the
cancer may be because near terminal patients with breast, colon, and ovarian
cancers may have been treated with many different chemotherapy regimens and
their cancers may have become chemotherapy resistant. In this way,
patients dying of chemotherapy responsive tumors may be more like decedents
with chemotherapy unresponsive cancers. This does not justify using
chemotherapy for unresponsive tumors. It also raises the question of
whether providing chemotherapy in the last 3 months of life to nearly a
quarter of cancer patients whose tumors have become resistant to chemotherapy
is itself an indication of overuse. This
study suggests that use of ineffective chemotherapy consumes substantial
medical resources. Annual expenditures for patients who received
chemotherapy, regardless of the responsiveness of the cancer, were 32.5%
higher than for patients who did not receive chemotherapy in the last 6 months
of life. Among patients who died of chemotherapy unresponsive cancers,
the use of chemotherapy in the last 6 months of life was associated with 20%
higher annual expenditures, or more than $5,500 per decedent. The extra
amount spent on providing chemotherapy to patients dying of unresponsive
cancers is comparable to the average annual expenditure for all Medicare
beneficiaries and nearly one third higher than annual per capita health
expenditures in the U.S. These data contrast with studies suggesting
that compared to "best supportive care" chemotherapy for non-small cell
lung cancer does not increase, and may even decrease medical costs.[xxi],[xxii],[xxiii]
The disjunction between our results and these studies may arise because of the
difficulty in translating results of randomized trials into actual clinical
practice. Care protocols in research may limit use of unnecessary
interventions, whereas in actual clinical practice use of treatments,
hospitalizations, and other interventions vary more. Furthermore, the cost
data on best supportive care come only from Canada and are more than a decade
old21-23, and patients receiving best supportive care were frequently
hospitalized, using more hospital days than patients receiving chemotherapy.
These old data, especially of hospitalizing patients receiving "best
supportive care" reflect practices not found in these data and unlikely to
still be common. It may also be that in actual clinical practice patients not
receiving chemotherapy may not be receiving "best supportive care"
reducing expenditures. Finally,
this overuse of treatment at the end of life is particularly wasteful when
placed in the context of the documented underuse of treatments proven by
randomized controlled trials to be effective in prolonging life. Studies
have shown that only 55% of Medicare beneficiaries receive adjuvant
chemotherapy for Stage III colon cancer.[xxiv]
Indeed, among 85 year old patients the use of chemotherapy for Stage III colon
cancer is 11% less than the frequency of the use of chemotherapy in the last 3
months for 85 year olds with chemotherapy unresponsive cancers.
Unfortunately, it appears that there may be overuse of chemotherapies in the
last few months of life coincident with underuse of therapies known to be
effective in prolonging life. In
health care, Massachusetts is known as a high use and high cost state. [xxv] A major issue is whether
these data on chemotherapy use at the end-of-life are unique to Massachusetts
or are generalizable. While there are some differences in the absolute
use of chemotherapy for some cancers, our data from California, although
limited, suggest a similar pattern of use of chemotherapy at the end of life.
In California one in five cancer decedents receive chemotherapy in the last 3
months of life, and this does not differ between chemotherapy responsive and
unresponsive cancers. Clearly, these results need to be confirmed
in other, larger populations. However, these data show that the situation in
Massachusetts is not unique. This
study has some significant limitations. First, the data may not
generalize in other ways. Chemotherapy use among decedents under 65 years of
age might be different. The strong trends toward greater use of chemotherapy
among younger decedents suggests these data might actually underestimate
chemotherapy use in the last 6 months of life among cancer decedents of all
ages. Chemotherapy use in managed care settings also might differ.
Second, we have no data on stage of cancer; some patients may have died from
acute toxicities of chemotherapy without being terminally ill. However,
data from trials suggest that acute toxic deaths among patients receiving
adjuvant therapy are rare, and thus unlikely to account for a substantial
proportion of cancer mortality.[xxvi]
Indeed, adjuvant chemotherapies associated with high toxic mortality would be
used infrequently. Third, the cause of death listed on death
certificates is not always accurate. However, listing cancer as the cause of
death may be insensitive, but it is specific, and Massachusetts and California
are among the states with the most accurate death certificates. Fourth,
annual expenditures were calculated but we tracked chemotherapy use only in
the last 6 months of life. Decedents who received chemotherapy in the 7 to 12
months before death only are classified in the "no chemotherapy" group,
increasing the costs of this group. This makes the difference in
expenditures appear smaller than if the comparison had been with decedents who
had received no chemotherapy in the entire last year of life. Most
importantly, these data provide no explanation for why chemotherapy is
provided in any particular case. Additional study is needed to determine
the reasons why chemotherapy is used in the last 6 months of life, especially
for chemotherapy unresponsive cancers. CONCLUSION
There
is substantial disagreement about whether chemotherapy is used appropriately
in patients near the end of life. This study demonstrates that one third
of patients in Massachusetts receive chemotherapy in the last 6 months of
life, even among those persons dying from chemotherapy unresponsive cancers.
Oncologists should reconsider the use of chemotherapy at the end of life. ACKNOWLEDGEMENTS We would like to thank Joan
Warren, Deborah Schrag, and Peter Bach for helpful advice and comments on the
project and manuscript. We would also like to thank many questioners at the
2001 Annual American Society of Clinical Oncology meeting for helpful
challenges.
TABLE 1: Characteristics of Massachusetts Cancer
Decedents by Receipt of Chemotherapy in the Last 6 Months of Life
* Includes all acute and
chronic leukemias, non-Hodgkin lymphomas, Hodgkin's disease, but excludes
multiple myeloma.
* Includes all acute and
chronic leukemias, non-Hodgkin lymphomas, Hodgkin's disease, but excludes
multiple myeloma.
* Includes all acute and
chronic leukemias, non-Hodgkin lymphomas, Hodgkin's disease, but excludes
multiple myeloma. TABLE 4: Expenditures in the
Last Year of Life for Massachusetts Cancer Decedents by Receipt of Chemotherapy
in the Last 6 Months of Life
* Includes all acute and
chronic leukemias, non-Hodgkin lymphomas, Hodgkin's disease, but excludes
multiple myeloma. TABLE 5: The Characteristics of California
Cancer Decedents by Receipt of Chemotherapy in the Last 6 Months of Life
* Includes all acute and
chronic leukemias, non-Hodgkin lymphomas, Hodgkin's disease, but excludes
multiple myeloma. REFERENCES [i] Annas GJ. The changing landscape of human experimentation: Nuremberg, Helsinki, and beyond. Health Matrix 1992;2:119-140. [ii] Sontag S. Illness as metaphor (New York:Farrar, Strauss, and Giroux 1978).
[iii] McKhann CF. A time to die (New Haven, Conn.: Yale University Press 1999) introduction and chapter 1. [iv] Miller M. Phase I cancer trials: a collusion of misunderstanding Hastings Center Report 2000;30(4):34-43. [v] Earle CC, Venditti LN, Neumann PJ, et al. Who gets chemotherapy for metastatic lung cancer? Chest 2000;117:1239-46. [vi] Holland JF, Frei E, Bast RC, et al. (eds.) Cancer Medicine 5th edition (Hamilton, Ont. Canada: B.C. Decker 2000) chapters 88, 96, 98, 101,105, 109.
[vii] Tannock IF, Boyer M. When is cancer treatment worthwhile? N Engl J Med 1990;323:989-990. [viii] Cullen MH, Billingham LJ, Woodroffe CM, et al. Mitomycin, ifosfamide, and cisplatin in unresectable non-small cell lung cancer: effects on survival and quality of life. J Clin Oncol 1999;17:3188-94. [ix] Cullen M. "Best supportive care" has had its day. Lancet Oncology 2001;2:173-5. [x] Ruckdeschel JC. Is chemotherapy for metastatic nonsmall cell lung cancer "worth it"? J Clin Oncol 1990;8:1293-1296. [xi] Slevin ML, Stubbe L, Plant HJ, et al. Attitudes to chemotherapy: comparing views of patients with cancer with those of doctors, nurses, and general public. BMJ 1990;300;1458-60. [xii] Yellen SB, Cella DF. Someone to live for: social well-being, parenthood status, and decision-making in oncology. JCO 1995;13:1255-1264. [xiii] Field MJ, Cassel CK. Approaching death: improving care at the end of life (Washington, D.C.: National Academy Press 1997) introduction. [xiv] Gough IR, Furnival CM, Burnett W. Patients attitudes to chemotherapy for advanced gastrointestinal cancer. Clin Oncol 1981;7:5-11. [xv] Ellis PA, Smith IE, Hardy JR, et al. Symptom relief with MVP (mitomycin C, vinblastine and cisplatin) chemotherapy in advanced non-small cell lung cancer. British J Cancer 1995;71:366-70. [xvi] Anderson H, Hopwood P, Stephens RJ, et al. Gemcitabine plus support care (BSC) v. BSC in inoperable non-small cell lung cancer-a randomized trial with quality of life as the primary outcome British J Cancer 2000;83:447-453. [xvii] Cunningham D, Pyrhonen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998;352:1413-18. [xviii] Coyle C, Crump M, Pintilie M, Oza AM. Does palliative chemotherapy palliate? Evaluation of expectations, outcomes, and costs in women receiving chemotherapy for advanced ovarian cancer. JCO 2001;19:1266-74. [xix] Burris HA, Moore MJ, Andersen J et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreatic cancer: a randomized trial. JCO 1997;15:2403-13. [xx] Wan GJ, Coune MA, Cella DF. The influence of personal expectations on cancer patients' reports of health related quality of life. Psycho-oncology 1997;6:1-11. [xxi] Jaakkimainen L, Goodwin PJ, Pater J, Warde P, Murray N, Rapp E. Counting the costs of chemotherapy in a National Cancer Institute of Canada randomized trial in nonsmall cell lung cancer. JCO 1990;8:1301-09. [xxii] Evans WK, LeChevalier T. The cost-effectiveness of navelbine alone or in combination with cisplatin in comparison to other chemotherapy regimens and best supportive care in stage IV non-small cell lung cancer. Eur J Cancer 1996;32A:2249-55. [xxiii] Berthelot JM, Will BP, Evans WK, Coyle D, Earle CC, Bordeleau. Decision framework for chemotherapeutic interventions for metastatic non-small cell lung cancer. JNCI 2000;92:1321-9. [xxiv] Schrag D, Cramer LD, Bach PB, Begg CB. Age and adjuvant chemotherapy use following surgery for state III colon cancer. JNCI 2001; (in press). [xxv] Center for the Evaluative Sciences The Dartmouth Atlas of Health Care (Chicago: American Hospital Publishing 1996). [xxvi] Fisher B, Brown AM, Dimitrov NV, et al. Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide, methotrexate, and fluorouracil in positive-node breast cancer patients with tamoxifen-nonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. JCO 1990;8:1483-96.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
