Near-Death Patients Often Accept Chemotherapy for Small Gains

Patients facing death from cancer are willing to undergo chemotherapy for much smaller improvements in outcome than would health care providers, a systematic review of the literature spanning two decades and multiple continents has found. A lack of decision aids and clear information about the benefits and risks of treatment may help explain why cancer patients often opt for aggressive care near the end of life, the researchers report in the Journal of Clinical Oncology (2006;24:3490—3496). At the same time, it is unclear how much better information about prognosis and treatment choices would alter the decision—making process, they said. “Our biggest finding is that we really need to learn a lot,”said lead author Robin K. Matsuyama, PhD, an instructor in internal medicine at Virginia Commonwealth University in Richmond. “Many studies seem to show that patients would choose chemotherapy even for a small benefit and even if there are side effects. But one big question is how many side effects they have experienced at that point. “There are so many aspects to the issue,”Dr. Matsuyama continued. “From the patient side, how much do they understand and how much do they want to know. From the physician side, how much do they want to say and how much are they comfortable saying. Would patients make the same choices as health care providers if they had all the same information? These are all questions that need further exploration.” End—of—Life Chemo on Rise For the review, the researchers searched PubMed from 1980 to 2005 for articles on “chemotherapy palliative care,”“medical decision—making,”“chemotherapy end—of—life,”“terminal illness treatment,”“decision chemotherapy treatment,”“prognosis communication,”“decision aids,”and “patient expectations.”More than 50 studies were identified. The researchers noted that the report comes at a time when the use of chemotherapy in the last weeks of life is on the rise. In 1996, 22% of Medicare patients started a new chemotherapy regimen in the last month of life, according to data cited in the review. Treatment within two weeks of death increased from 13.8% in 1993 to 18.5% in 1996 (Earle C et al: JCO 2004;22:315—321). In that study, the researchers found that chemotherapy use was similar for patients who were considered responsive and unresponsive to the drugs, Dr. Matsuyama said. Additionally, an unpublished 2005 study showed that oncology practices administered chemotherapy to anywhere from 0% to 34% of patients in their last two weeks of life, the researchers said. All of the available studies support the notion that patients with cancer make different choices about treatment than would other patients or the health professionals caring for them, Dr. Matsuyama said. One study showed that patients from the United Kingdom were far more willing to undergo hypothetical toxic cancer treatment than were their doctors and nurses—a finding that remained true after chemotherapy (BMJ 1990;300:1458—1460). In another trial, Norwegian patients were significantly more willing to accept a toxic chemotherapy regimen compared with oncologists, surgeons, and controls (Eur J Cancer 1995;31A:1917—1918). Yet another study showed that Japanese patients with lung cancer were significantly more likely than other patients with respiratory diseases to accept treatments for a potentially small benefit (Intern Med 2005;44:107—113). The list goes on. Other research points to a variation in individual patient responses. In one US study, some patients with previously treated non—small—cell lung cancer said they would accept chemotherapy for a survival benefit of one week, while others would not undergo chemotherapy even if it added two years to their lives (BMJ 1998;317:771—775). The findings challenge the common suggestion that oncologists overprescribe chemotherapy near the end of life or force patients to accept chemotherapy against their wishes, Dr. Matsuyama said. “There's always been a universal sense that physicians push chemotherapy, but many patients are asking for treatment. Patients value even small benefits greatly and believe that toxicity is less important than even the hope of small gains. Oncologists may be the most cognizant of how hopeful patients are for small benefits.” The Patient Perspective One of the most startling findings of the review was that patients learn more about the progression of disease and their chance of death from fellow patients in the waiting room than from their oncologists, Dr. Matsuyama said. The observation comes from the only study of its kind, she said, in which 35 small—cell lung cancer patients were observed to determine their sources of information and knowledge about their disease. The researchers found that physicians did not always want to pronounce a “death sentence,”and patients did not always want to hear it (West J Med 2001;174:247—253). Other studies showed that patients may not believe or understand the information being presented to them, the researchers reported. One study showed that one third of metastatic lung cancer patients thought that they were receiving therapy with curative intent (Br J Cancer 1998;58:355—358), while another study showed that 35% of patients believed their palliative radiation represented a cure (Clin Oncol [R Coll Radiol] 2001;13:204—208). The Communication Gap Other research revealed that physicians do not always accurately predict a patient's prognosis or communicate it to patients. A meta—analysis of 13 studies, for example, showed that oncologists consistently overestimate prognosis by at least 30% (BMJ 2003;327:195—200). Another study showed that physicians overestimate survival by a factor of 5.3 (BMJ 2000;320:469—473). Yet other researchers found that physicians would provide cancer patients with survival estimates in only 37% of cases—even if the patients requested them (Ann Intern Med 2001;134:1096—1105). A rosy prognosis may not be harmful in itself, but it is associated with detrimental outcomes, Dr. Matsuyama and colleagues noted. In the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment (SUPPORT), patients who overestimated their prognosis had the same survival, but a greater chance of dying in the intensive care unit, while incubated, or with resuscitation, compared with other patients (JAMA 1998;279:1709—1714).figure: Susan Block, MD: “Physicians who want to avoid sharing a poor prognosis to minimize patient suffering may be doing patients a disservice.”The vast majority of patients say they want their oncologists to lay out the facts about their prognosis, the review showed. In one recent study of 126 patients, 98% said they wanted their oncologists to be realistic, provide an opportunity for them to ask questions, and acknowledge them as an individual when discussing prognosis (JCO 2005;23:1278—1288). Decision aids can help patients understand their options and guide patient choice, the review revealed. In a randomized controlled trial, only 58% (35 of 60) of breast cancer patients who were provided with a decision aid with individualized risk information chose adjuvant chemotherapy, compared with 87% (33 of 38) of their counterparts in usual care (Med Decis Making 2005;25:301—307). Clear Information Lacking But obtaining honest and useful information about prognosis and treatment options is not easy, Dr. Matsuyama said. For example, she and her coauthors pointed out, on the NCI's Web site (www.cancer.gov), the section on non—small—cell lung cancer lists treatments, but does not provide the likelihood that any of the approaches will help the individual patient. There is also no information about prognosis, the benefits and risks of first— and second—line chemotherapy, or the impact of treatment on quality of life. Similarly, Dr. Matsuyama continued, the Lung Cancer Alliance Web site (www.lungcanceralliance.org) offers no information about prognosis for individual patients, the expected outcomes of treatment, what treatments work and how well, the alternative of palliative care without specific anticancer treatment, how to choose among therapies, or what most patients could reasonably. Neither, Dr. Matsuyama noted, are there decision aids proven to assist patients with metastatic disease in making their choices. “Decision aids for advanced lung cancer patients and methods such as the Ottawa Decision Support Framework hold promise but have not been adopted in practice,”she said. ‘Complexity of the Issue’ Asked for her opinion, Susan Block, MD, Chief of the Division of Psychosocial Oncology and Palliative Care at Dana—Farber Cancer Institute, said that the researchers did a “really nice job dissecting the complexity of the issue as well as the gaps in what we know.” The finding that patients would opt for chemotherapy for much smaller improvements in outcome than would health care providers may lead some to conclude that oncologists should not impose their opinions on patients because of their different perspectives, she noted. “But really, we are not so different from patients. That leads to the question of whether patients would make different choices if they had all the information we as oncologists have.” Physicians who want to avoid sharing a poor prognosis to minimize patient suffering may be doing patients a disservice, Dr. Block said. “The patients are not being prepared for the reality that their life will end.” Next Step So what should oncologists be doing? “We have to explore with patients how much information they want about their illness and how they anticipate that information will help them to live,”she said. Dr. Block added that in her experience, an overwhelming number of patients want all the information available. While oncologists may be reluctant to be pinned down “to a time frame that is often uncertain, patients can deal with uncertainty,”she said. “Make an estimate,”she urged other oncologists. “Say, ‘My hunch is that you have a week or a small number of months.’ That kind of communication can be very helpful to patients in planning and prioritizing and can help them to make better decisions about treatment.” Craig C. Earle, MD, MSc, Associate Professor of Medicine at Harvard Medical School and a researcher specializing in quality of care and outcomes for cancer patients at Dana—Farber Cancer Institute, agreed. In an editorial accompanying the study, Dr. Earle said, “We have all encountered patients who prefer not to know anything about their prognosis or refuse to believe it once it is conveyed. However, surveys suggest that the majority of patients do want an accurate estimate of prognosis, even if it is poor. “Well—intentioned mistruths can actually be detrimental to patients,”he continued. “Data are presented in the article that patients with unrealistic expectations were more likely to choose aggressive treatment, but lived no longer and were more likely to die in intensive care on life support. An overly rosy prognosis deprives patients of the ability to make truly informed decisions about whether to take chemotherapy rather than a final, important, and meaningful trip.”figure: Craig C. Earle, MD, MSc: “Surveys suggest that the majority of patients do want an accurate estimate of prognosis, even if it is poor.”Education, Decision Aids That said, information about prognosis may or may not change an individual patient's behavior, which, as the review found, can vary widely, Dr. Block noted. “For one patient, a one in 100 shot at more time may be worth it, while another patient may want to avoid further chemotherapy no matter what.” Dr. Matsuyama stressed the importance of patient education. “We have to develop some really clear information and decision aids to help patients understand their illness and treatment options,”she said, adding that her group has applied for grants to develop decision—making tools. But Dr. Earle said that decision aids are not enough. “Decision aids can help impart facts, but there is also a need to fulfill our role as expert advisers,”he said. “We must combine interpretation of objective data with clinical judgment about a particular patient's situation and not be afraid to make a recommendation about what we think will be best for them.”

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.