|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.06.050 on November 1 2004 © 2004 American Society of Clinical Oncology. Perspectives, Preferences, Care Practices, and Outcomes Among Older and Middle-Aged Patients With Late-Stage CancerFrom the Case Western Reserve University School of Medicine; MetroHealth Medical Center; Louis Stokes Cleveland Veterans Affairs Medical Center (VAMC) Geriatric Research Education and Clinical Center (GRECC); Center for Health Care Research and Policy, Cleveland, OH; Beth Israel Deaconess Medical Center, Boston, MA; Duke University School of Medicine, VAMC, Durham, NC. Address reprint requests to Julia Hannum Rose, PhD, MA, Department of Medicine–Geriatrics, Case Western Reserve University, 2500 MetroHealth Dr, Cleveland, OH 44109; e-mail: julia.rose{at}case.edu
PURPOSE: To evaluate relationships among physician and cancer patient survival estimates, patients' perceived quality of life, care preferences, and outcomes, and how they vary across middle-aged and older patient groups.
PATIENTS AND METHODS: Subjects were from the Study to Understand Prognoses and Preferences for Risks of Treatments (SUPPORT) prospective cohort studied in five US teaching hospitals (from 1989 to 1994), and included 720 middle-aged (45 to 64 years) and 696 older ( RESULTS: Although most patients had treatment goals to relieve pain, treatment preferences and care practices were linked only in the older group. For older patients, preference for life-extending treatment was associated with more therapeutic interventions and more documented discussions; cardiopulmonary resuscitation (CPR) preference was linked to more therapeutic interventions and longer survival. For middle-aged patients, better perceived quality of life was associated with preferring CPR. In both groups, patients' higher survival estimates were associated with preferences for life-prolonging treatment and CPR; physicians' higher survival estimates were associated with patients' preferences for CPR, fewer documented treatment limitation discussions about care, and actual 6-month survival. More discussions were associated with readmissions and earlier death. More aggressive care was not related to outcomes. CONCLUSION: Fewer older patients preferred CPR or life-prolonging treatments. Although older patients’ goals for aggressive treatment were related to care, this was not so for middle-aged patients. Aggressive care was not related to prolonged life in either group.
As our society ages, increasing numbers of middle-aged and older people living in the United States will be diagnosed with, and will eventually die of cancer.1-3 A diagnosis of incurable cancer at different stages of the adult life span holds different meanings for patients, health care providers, and society. Although patient autonomy in medical decision making is widely supported in this society, previous research has demonstrated that treatment decisions are often made with insufficient attention to patients' perspectives and wishes,4-7 and that age is an important consideration.8-11 Studies suggest that cancer patients' and/or their physicians' perspectives may vary, in part depending on patient age.12-17 Age-related differences in decision making about medical care have been found in both hospital- and community-based settings.3,18-22 Physicians' perspectives are strongly related to treatment decisions even when patients' perspectives are not. Under the concept of informed consent, the degree to which patients may exercise "autonomous" decision making depends on input from physicians. Physicians involved in near end-of-life care may be able to provide more useful input by better understanding both the effect of age on patients' preferences and perspectives, and its relationship to current practices and outcomes. Support for patients and families making decisions may also be enhanced by an improved understanding of this process.
Although older (age
Such studies, however, have not been conducted with hospitalized patients with late-stage cancer. It is important to note that older patients receiving hospital care in advanced cancer are, largely, "young-old" (65 to 84 years) rather than "old-old" adults ( Whereas some patients with advanced cancer opt for nonhospital care, hospitalized patients generally represent those who are most likely to receive some degree of aggressive care.8,18,33 Moreover, the relationships among perspectives, preferences, and cancer-related treatments or procedures received within approximately the first week of hospitalization are particularly salient because important treatment decisions are often made during this period.34 Documented discussions about treatment limitation during hospitalization represent an additional important aspect of care processes during the late stage of cancer in adults.6,35 To date, findings indicate that hospitalized, seriously ill older patients receive more documented discussions about treatment limitation and less aggressive care than do younger adult patients.8,18,19 What is less well understood is the effect of age on late-stage cancer patients' preferences and perspectives, and its relationship to current practices and outcomes. To our knowledge, this is the first large empirical study to (1) compare preferences and perspectives among hospitalized middle-aged and older late-stage cancer patients and (2) investigate the value of preferences and perspectives in the evaluation of care practices and outcomes. Figure 1 summarizes the conceptual model guiding the research and analysis. This model extends previous work, including findings reported on the Study to Understand Prognoses and Preferences for Risks of Treatments (SUPPORT) data.18,33,34 As shown in Figure 1, decision making about near end-of-life care is affected by many factors, including patient and physician perspectives on patients' likelihood of survival,5,6 as well as patients' perceived quality of life.7,36 Late-stage cancer patients' preferred treatment goals (to extend life or to relieve pain/discomfort), cardiopulmonary resuscitation (CPR), and related specific interventions are expected to influence care practices, especially in acute care settings. Those care practices include both documented discussions about aggressiveness of care and decisions about therapeutic interventions.
To better understand the relationships among perspectives, preferences, care practices (or actions), and outcomes, we pose two questions. (1) Within age groups, do differences exist (A) in the perspectives of patients and their physicians, or (B) in the care preferences of patients? (2) Within age groups, what are the patterns of association (A) between patient and physician perspectives and patient care preferences, and (B) among patient and physician perspectives, patient preferences, care practices, and outcomes? In our conceptual framework, we make an important distinction between perspectives and care preferences, and from an analytical standpoint, they will be treated as distinct variables. The models used in our analyses specify a series of regressions using specific preferences, care practices, and outcomes as dependent variables in a series of analyses within each age group.
Patients The analyses assessed middle-aged (45 to 64 years) and older (  65 years) adult cancer patients from a subset of the SUPPORT data. The SUPPORT study examined the care of seriously ill hospitalized patients, and was conducted in five US teaching hospitals (1989 to 1994). The SUPPORT project presented a unique opportunity to examine prospectively large numbers of middle-aged and older hospitalized late-stage cancer patients' and their physicians' perspectives, care practices, and outcomes.33 The median life expectancy for patients entering SUPPORT was estimated to be 6 months. Of the 9,105 patients enrolled in nine disease categories, 2,182 (24%) were cancer patients who had an admitting diagnosis of non–small-cell lung cancer, colon cancer with metastases to the liver, or malignancy associated with multiorgan system failure. Subjects in this report are cancer patients in the middle-aged and older adult age groups or their surrogates who completed an interview on study day 3. There were 696 older and 720 middle-aged patients enrolled. Findings from the SUPPORT project have shown that older patients were less likely to receive cancer-related treatments (eg, chemotherapy, platelet infusions, and scheduled intravenous medications) than middle-aged patients during the first week of hospitalization.18 Previous work also indicated that older patients had more documented discussions about treatment limitation than middle-aged patients. However, it is important to note that even though older patients had more discussions about transfer to hospice, they were no more likely to be discharged to hospice, or home with hospice or home care.18
Procedures In-depth interviews were conducted with study patients whenever possible. The condition of the study population of interest required the use of surrogate data to avoid bias created by missing data about incapacitated patients.39 Consistent with other studies, a surrogate decision maker was interviewed if the patient was unable to participate. Surrogates in SUPPORT served as primary decision makers about care practices and as spokespersons for care preferences of such patients. This reflects actual clinical decision-making practice, as surrogates are presumed to be the most informed and appropriate sources of preferences available for incapacitated patients. When patient data about preferences and perspectives were not available, surrogate data were substituted, when available. The subset of cases for which both patient and surrogate responses were available was used to calibrate the surrogate responses to obtain a distribution similar to that of patients' scores. When neither patient nor surrogate responses were available, a multiple imputation strategy was used.40
Measurement
Adjuster variables (listed in Table 2) in all the models were selected based on research on cancer and aging1,12,19 and previous work by SUPPORT investigators18,33,34 regarding factors that could potentially influence care practices and outcomes. For example, the SUPPORT prognostic model estimating the probability of surviving 2 months following study entry (ie, after hospital admission) was included as an adjuster for severity of illness during hospitalization.43 Variables for site of care (to adjust for practice variation among sites) and intervention status were also included. In the SUPPORT project, the intervention was not found to significantly affect outcomes,33 so phase I and phase II data are combined here.
Statistical Analysis All analyses were performed using SAS (SAS Institute, Cary, NC) software. To address differences between the age groups as far as patient and physician perspectives, patient preferences, care practice, and outcomes, we used  2 and Wilcoxon tests. We conducted separate analyses within age groups in order to evaluate the patterns of inter-relationships of these variables and whether, for each age group, they approximated or diverged from the conceptual models. Our strategy to conduct a separate analysis was chosen in order to provide more detailed and clinically relevant data that could inform care practices, specifically for middle-aged or older patients. In within-group analyses, we used multiple linear regression for continuous variables, logistic regression for dichotomous variables, and Cox proportional hazards for mortality44 (6-month timeframe; PHREG procedure, SAS Institute). Because of the potential relevance of "don't know" responses for treatment goals that is conceptually distinct from choosing to extend life or to relieve discomfort, we conducted polytomous logistic regressions, as appropriate. For clarity in explanation, we also conducted logistic regressions recoding don't know categories as more aggressive care, because in clinical practice, the default approach for treatment in the absence of a comfort care direction is to provide more aggressive care. In addition, models for readmission took into account censoring (ie, those who survived longer had a greater likelihood of being readmitted) by using the Anderson-Gill model.45,46 Models for survival included a binary measure of readmission as an additional adjuster given the relationship between readmission and survival. Follow-up analyses (logistic regressions) were conducted for two of the individual items (resuscitation and transfer to hospice) from the scale assessing documented discussion topics, because of their particular importance and demonstrated differences in frequency between middle-aged and older patients.18
Baseline Characteristics by Age Group Older and middle-aged patients were similar on most variables, except older patients had fewer years of education and had lower incomes, as expected. The mean age for the older patient group ( 65) was 72.4 years; only 3% of older patients were 85 years of age. For the middle-aged group (range, 40-64 years) the mean age was 56.9 years (Table 2).
Perspectives and Care Preferences by Age Group
Age group comparisons for care practices and outcomes are also included in Table 3, and confirm previously reported findings on these outcomes for this subgroup.8,18 Indeed, there was a trend for older adults to receive fewer therapeutic interventions during the first week of hospitalization, and the number of topics discussed about aggressiveness of care was greater among older patients. A larger proportion of middle-aged patients was readmitted to the index hospital and survived to 6 months.
Patterns of Association Within Age Groups
Middle-aged patients. Among middle aged patients, a preference to extend life (or uncertainty about this choice), versus a clear choice for comfort care, was significantly related to the patient's higher estimate of survival. However, it was not related to the physician's prognostic estimate. Perceived quality of life was unrelated to a preference to extend life. The more specific preference for CPR was associated with both physician and patient estimates of survival. Middle-aged patients who thought of the quality of their lives as poorer were less likely to prefer CPR. As expected, a preference for CPR was positively associated with a preference to extend life. Regarding care practices, the number of topics discussed about aggressiveness of care was related to the physician's poorer prognosis for survival. Neither discussions about care nor the number of therapeutic interventions was associated with any patient preference or perspective. Both readmission and earlier death were related to the number of topics discussed regarding treatment limitations and aggressiveness of care. Readmission was not predicted by any patient preferences, by patient perceived quality of life, or by patient or physician prognostic estimates for survival. Survival to 6 months was associated with patient and physician prognostic estimates, but not with patient-perceived quality of life or with any patient preference. There was no association between the number of therapeutic interventions and the length of survival for patients in this age group. Older patients. Among patients 65 years and older, preference to extend life was again related to the patient's estimate for survival, but was not associated with the physician prognostic estimate or with perceived quality of life. Specific preference for CPR was associated with both prognostic estimates for survival, and, as expected, was positively associated with a preference to extend life, and inversely related to a preference for relief of pain or discomfort. Older patients' perceived quality of life was not associated with care preferences. The number of topics discussed regarding aggressiveness of care was associated with a poorer prognosis by the physician, and to the patient's preference to extend life. Older patients who preferred pain relief to extending life discussed fewer topics with their physicians. The number of therapeutic interventions increased with physician estimate for survival and with patient preferences for CPR and for treatments to extend life. Among older patients, readmission and death were associated with the number of topics discussed regarding treatment limitations and aggressiveness of care. Patient preference for CPR was associated with survival to 6 months. Follow-up analyses. Additional within–age-group analyses were done on two discussion topics about treatment limitation: resuscitation and transfer to hospice. In both age groups, poorer prognostic estimates by physicians were related to having documented discussions about resuscitation (P < .001 for both groups) and discussions about transfer to hospice (P < .01 and P < .001 for middle-aged and older adults, respectively). In older patients, a preference for no CPR was associated with discussions about both topics (P < .05). Also, older patients who said they did not know versus those who voiced a preference for treatment to relieve pain and discomfort, were more likely to have documented discussions about resuscitation (P < .05). In middle-aged patients, although patient preferences for care were not associated with such discussions, poorer perceived quality of life and greater pessimism about survival were associated with discussion about transfer to hospice (P < .05). Figures 3 and 4 provide a conceptual schemata to describe results of within-group analyses shown in the Tables.
Our research was designed to explore age group similarities and differences in patient and physician perspectives and to better understand the relationship between perspectives and patient preferences for treatment, including CPR wishes. Previous research has shown that patterns of decision making and care for patients diagnosed with incurable cancer varies with age. Most importantly, within each age group, we examined specific patterns of association among physician and patient perspectives, patient preferences, care practices, and outcomes. Our conceptual model represents the premise that communication is the fundamental mechanism for connection between perspectives and practices in medical decision making for end-stage cancer patients. In an ideal end-of-life communication scenario, patient preferences and perspectives should be linked to specific actions, such as CPR, discussions about aggressiveness of care, and therapeutic interventions. (It is important to note that other measures of severity and burden of chronic illness were included in these analyses to adjust for the potential confounding between use of surrogate and health status.) Although the link between patient and physician perspectives and patient care preferences was found in both age groups, the link between patient preferences and actual care practices, and outcomes as well, was found only in the older age group. The absence of such associations in middle-aged patients is a new discovery and an important finding, especially in light of reports that middle-aged patients desire greater involvement in health care decision making than the current cohort of older adults.9,15,47 An additional important finding was the lack of association between aggressiveness of care and prolongation of life as an outcome for patients in either age group. In both age groups, although more than half of the patients preferred comfort care as a treatment goal, the majority would still opt for the specific option of CPR. From a clinical perspective, these results may seem inconsistent. It may be that patients do not have a clear understanding of the implications of life-extending treatments, or that their responses reflect a variety of factors other than clinical logic. Perhaps it is easier for patients, regardless of age, to choose nonaggressive care as a broad treatment goal than to emphatically decline resuscitation, with its immediate life or death consequence. Patients also may not understand the issues fully, including the clinical realities of invasive procedures. The finding that a larger percentage of older than middle-aged patients said they did not know their treatment goal will require further study to explain. It is possible that, as opposed to being undecided, more older patients were ambivalent about the mutually exclusive treatment goals presented to them, and preferred care practices that lie on a continuum of care somewhere between these two extremes. Alternately, older patients may prefer stepped, situation-specific decisions, rather than global choices. It is also possible that, given the general trends of somewhat poorer quality of life and less optimistic survival estimates among these older patients compared with their younger counterparts, life-extending treatments, and care to relieve pain may have a different meaning for this group. Results of this study provide further evidence that adult patients with advanced cancer tend to be overly optimistic about the likelihood of survival into the future.5,6 The finding that older patients were only slightly less optimistic about survival than middle-aged patients may, in part, reflect the unique attributes of older patients who seek acute care under these circumstances. However, similar levels of optimism also have been found in comparable nonhospitalized older cancer patients,6 indicating that this phenomenon is more widespread. It is important to note that older subjects in both settings were "young-old," which may be an important consideration for generalization. In contrast to patients, physicians were much less optimistic about patients' survival in both age groups, especially older patients. Physicians are known to be more accurate than cancer patients,48 raising questions about the need to communicate more effectively with patients5,35,49,50 and to better understand the specific perceptions and coping styles of patients (eg, monitoring v blunting51,52) in both age groups. It was anticipated that physicians' and patients' perspectives on patients' likelihood to survive 6 months or more and patients' own perception of quality of life would be salient factors in patients' preferences for treatment. Instead, only the patients' survival estimates were associated with treatment goals of patients in both age groups. This finding highlights the importance of understanding the patient's perspective on his or her own survival while conducting discussions about aggressiveness of care. Much of the literature about end-of-life decision making addresses decisions about whether or not CPR should be performed. There is consensus that these decisions should be tied to patient preference and clinical realities.4,53 In both age groups, physician and patient estimates for patient survival and patients' general preference for treatment to extend life were associated with patients' preference for CPR, underlining the important role of expectations for survival in making this choice. A better perceived quality of life also predicted preference for CPR in the middle-aged, and there was a similar trend in the older age group. Quality of life was expected to be an important consideration in both age groups32,36 and its modest impact may be due in part to how it was measured (ie, by a single item). Although the majority of patients in both age groups had treatment goals of relieving pain or discomfort, the link between preference for such treatment and care practices was found only in the older age group. This finding raises the question whether the wishes of older patients who prefer nonaggressive care are more readily accepted in care decision making than the wishes of middle-aged patients who have similar preferences. By the same token, older patients who clearly preferred care to extend life, including CPR, received more therapeutic interventions and more documented discussions about treatment limitation. Indeed, older patients wanting treatment to prolong life may need to voice their preferences, possibly countering stereotypes against a more aggressive approach in the elderly. This may pose a dilemma for the current cohort of older patients who tend to be less communicative or involved in care decision making.9-11,47,54 In contrast, among the middle-aged, more aggressive forms of treatment may be routinely pursued regardless of patients' preferences. The finding of no association between patient preferences and care practices among this group is of concern, especially given greater desire among the middle-aged for involvement in care decision making,9,15,47 and may illustrate a possible disparity in practice that warrants further investigation. Although more documented discussions were conducted with older patients who did not clearly prefer nonaggressive care (ie, those who did not know their treatment goal or preferred treatment to extend life), similar preferences by middle-aged patients did not prompt more documented discussions about aggressiveness of care. The unique finding that middle-aged patients' greater pessimism about survival and poorer perceived quality of life were tied to discussion about transfer to hospice raises the question whether patient readiness is especially important in approaching this topic with younger patients. These observed age-related differences have implications for improving communication about prognosis and care preferences in late-stage cancer, perhaps especially for middle-aged patients and their families, to better meet the goals of patient-centered decision making.53,54 The finding that neither preferences nor therapeutic interventions predicted readmission of patients was unexpected. Instead, after adjusting for severity of illness, more documented discussion about treatment limitation was the single predictor of readmission to the index hospital in both age groups. In this study, although discussions were documented in terms of aggressiveness of care topics, the actual content of communication is not known. Thus, it is unclear the extent to which these discussions may have focused on receiving or avoiding aggressive care into the future. However, because the majority of patients died in the hospital,18 and such discussions also were associated with earlier death (potentially during readmission), the focus was likely to have been on treatment limitation. In addition to more documented discussions about treatment limitation, more pessimistic prognostic estimates by physicians and patients alike were associated with earlier death in both age groups. In general, sicker patients as well as their physicians appeared to recognize what lay ahead. The finding that older patients' preference for CPR was associated with longer survival, controlling for severity of illness, may indicate that personal determination or investment in living plays an important role, especially in older patients. This study shows that both middle-aged and older cancer patients and their physicians have very different perspectives on the likelihood of survival by patients with advanced cancer. Given an expectation that choices about aggressiveness of care are best made based on patient preferences informed by clinical realities, the disparities in perceptions about survival present an opportunity for improvement in physician-patient communication about prognosis. These findings also illustrate the possibility that patients' preferences for care may have little, if any, bearing on hospital readmissions among middle-aged and older cancer patients. The result that patient preference for CPR is related to improved survival only among older patients may indicate differential care by age –a topic worth examining in clinical practice. These findings, combined with previous reports,5,6,35,53-55 suggest that improved communication regarding prognosis and near end-of-life preferences for care is warranted, and that age is an important consideration. Of particular interest is the finding that patient preference for treatments to extend life does not appear to translate into a survival advantage, although among older patients this preference is related to a greater number of interventions for that purpose. Are there benefits other than survival that we are not measuring? The success of the decision-making partnership between patient and health care provider is rooted in ongoing communication between the partners about crucial information including diagnosis, prognosis, treatment options, and potential outcomes. Both middle-aged and older patients must be encouraged to assess how these options relate to their own coping and values and to convey their preferences, needs, and goals.51-54 The classic model of such decision-making, a benefit-burden analysis, is contingent on exchange that is both age sensitive and unbiased. Indeed, our analyses were prompted by previous work indicating that cancer patient and physician perspectives may vary in part depending on patient age,12-17 and that age-related differences exist in medical decision making about care in both hospital and community–based settings.3,18-22 Our primary focus was on examining relationships between perspectives, preferences, and practices in aggressiveness of care for middle-aged and older hospitalized late-stage cancer patients. Most importantly, as discussed above, our analyses provide more detailed and clinically relevant data that may inform efforts to improve care practices targeted for patients within each age group. Although we were looking for indications that hospitalized older patients may be more overlooked in decision making, that is not what we found. Instead, we report an apparent disconnect between preferences of middle-aged patients and the aggressiveness of care they received. This is an important and troublesome finding when one takes into account previous reports that hospitalized middle-aged patients tend to have fewer discussions about treatment limitation, and higher resource utilization, than comparable older patients.8,18 This is especially true when one considers that over half of middle-aged patients in our analyses reported that they preferred a treatment goal to relieve pain or discomfort over extending life, or remained unsure. If patient preferences and physician prognostic estimates do not affect aggressiveness of care received in hospitalized middle-aged patients, what does? In order to pursue this and other important questions posed in our discussion, more complex models must be developed for conducting within age group analysis. Models for middle-aged and older groups may differ and will require the inclusion of additional key variables, as well as the use of potentially more valid and reliable measures to assess perspectives, preferences, care practices, and outcomes. At a minimum, patient and family coping and communication style, physician perspectives and preferences in cancer care decision making, and broader health system and social factors56 must be considered in future studies.
The authors indicated no potential conflicts of interest.
We acknowledge Joanne Lynn, Russell Phillips, and Neil Wenger for helpful editorial comments.
Supported by National Cancer Institute grant RO-1 CA72977-01 (J.H.R.). The SUPPORT Project was funded by the Robert Wood Johnson Foundation. The opinions in this manuscript do not necessarily represent the views of the Foundation. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Yancik R: Cancer burden in the aged: An epidemiological and demographic overview of cancer. Cancer 80:1261-1266, 1997[CrossRef][Medline] 2. Merrill SS, Verbrugge LM: Health and disease in midlife, in Willis SL, Reid JD (eds): Life in the middle. San Diego, CA, Academic Press, 1999, pp 77-103 3. Balducci L, Beghe C: Management of cancer in the older person. Clin Geriatr 10:54-60, 2002 4. Danis M, Southerland LI, Garrett JM, et al: A prospective study of advance directives for life-sustaining care. N Engl J Med 324:882-888, 1991[Abstract]
5. Weeks JC, Cook FC, O'Day SJ, et al: Relationship between cancer patients' predictions of prognosis and their treatment preferences. JAMA 279:1709-1714, 1998
6. Fried TR, Bradley EH, Towle VR, et al: Understanding the treatment preferences of seriously ill patients. N Engl J Med 346:1061-1066, 2002
7. Lynn J, Teno JM, Phillips RS, et al: Perceptions of family members of the dying experience of older and seriously ill patients. Ann Intern Med 126:97-106, 1997
8. Hamel MB, Teno JM, Goldman L, et al: Patient age and decisions to withhold life-sustaining treatments for seriously ill, hospitalized adults. Ann Intern Med 130:116-125, 1999 9. Beisecker AE: Aging and the desire for information and input in medical decisions: Patient consumerism in medical encounters. Gerontol 28:330-335, 1998 10. Coe RM, Miller DK. Communication between hospitalized older patient and physician. Clin Geriatr Med 16:109-118 11. Siegler EL, Levin BW: Physician-older patient communication at the end of life. Clinics Geriatr Med 16:175-204, 2000
12. Yancik R, Ganz P, Varricchio G, et al: Perspectives on comorbidity and cancer in older patients: Approaches to expand the knowledge base. J Clin Oncol 19:1147-1151, 2001 13. Ganz PA: Interaction between the physician and the older patient: The oncologist's perspective. Cancer 80:1323-1325, 1997[CrossRef][Medline] 14. Silliman RA, Troyan SL, Guadagnoli E, et al: The impact of age, marital status and physician-patient interactions on the care of older women with breast carcinoma. Cancer 80:1326-1333, 1997[CrossRef][Medline] 15. Blanchard CG, Labrecque MS, Ruckdeschel JC, et al: Information and decision-making preferences of hospitalized adult cancer patients. Soc Sci Med 27:1139-1145, 1988
16. Hofmann JC, Wenger NS, Davis RB, et al: Patient preferences for communication with physicians about end-of-life decisions. Ann Intern Med 127:1-12, 1997 17. Harrison J, Maguire P: Influence of age on psychological adjustment to cancer. Psycho-Oncol 4:33-38, 1995 18. Rose JH, O'Toole EE, Dawson NV, et al: Age differences in care practices and outcomes for hospitalized patients with cancer. J Am Geriatr Soc 48:S25-S32, 2000 (suppl)[Medline]
19. Earle CC, Tsai JS, Gelber RD, et al: Effectiveness of chemotherapy for advanced lung cancer in the elderly; Instrumental variable and propensity analysis. J Clin Oncol 19:1064-1070, 2001 20. Greenfield S, Aronow HU, Ganz PA, Elashoff RM: The effect of age in the management of elderly cancer patients, in Yancik R, Yates J (eds.) Cancer in the elderly. New York, NY, Springer, 1989, pp 55-69 21. Hamel MB, Phillips RS, Teno JM, et al: Seriously ill hospitalized adults: Do we spend less on older patients? J Am Geriatr Soc 44:1043-1048, 1996[Medline] 22. Cleary JF, Carbone PP: Palliative medicine in the elderly. Cancer 80:1335-1347, 1997[CrossRef][Medline] 23. Ryff CD, Essex MJ: Psychological well-being in adulthood and old age: Descriptive markers and explanatory processes, in Schaie WE (ed). Annual review of gerontology and geriatrics. New York, NY, Springer, 1999, pp 144-171 24. Staudinger UM, Bluck S: A view of middle life development from lifespan theory, in Lachman ME (ed): Handbook of Midlife Development. New York, NY, Lawrence Earlbaum, 2001, pp 189-206 25. Merluzzi TV, Nairn RC: Adulthood and aging: Transitions in health and health cognition, in Merluzzi TV (ed): Life-Span Perspectives on Health and Illness. Mahwah, NJ, Lawrence Erlbaum, 1999, pp 44-57 26. Rose JH: A life course perspective on health threats in aging. J Gerontol Soc Work 17:85-97, 1991[CrossRef] 27. Mages NL, Mendelsohn G: Effects of cancer on patients' lives: A personological approach, in Stone GC, Cohen F, Adler NE (eds). Health Psychology: A Handbook. San Francisco, CA, Jossey-Bass, 1979, pp 255-285 28. Ganz PA, Schag CC, Heinrich RL: The psychosocial impact of cancer on the elderly: A comparison with younger patients. J Am Geriatr Soc 33:429-435, 1985[Medline] 29. Mor V, Allen S, Malin M: The psychosocial impact of cancer on older versus younger patients and their families. Cancer 74:2118-2127, 1994 (suppl 7)[CrossRef][Medline] 30. Prohaska TR, Funch D, Blesch KS: Age patterns in symptom perception and illness behavior among colorectal cancer patients. Behav Health Aging 1:27-39, 1990 31. Rose JH: Interactions between patients and providers: An exploratory study of age differences in emotional support. J Psychosoc Oncol 10:43-67, 1993[CrossRef] 32. McCarthy EP, Phillips RS, Zhong Z, et al: Dying with cancer: Patients' function, symptoms and care preferences as death approaches. J Am Geriatr Soc 48:110-121, 2000 (suppl 5) 33. SUPPORT Principal Investigators: A controlled trial to improve care for seriously ill hospitalized patients: The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). JAMA 274:1591-1598, 1995[Abstract] 34. Rose JH, O'Toole EE, Dawson NV, et al: Generalists and oncologists show similar care practices and outcomes for hospitalized late-stage cancer patients. Med Care 38:1103-1118, 2000[CrossRef][Medline] 35. Bradley EH, Hallemeier BA, Fried TR, et al: Documentation of discussion about prognosis with terminally ill patients. Am J Med 111:218-223, 2001[CrossRef][Medline]
36. Singer PA, Martin DK, Kelner M: Quality of end of life care: Patients' perspectives. JAMA 281:163-168, 1999 37. Murphy DJ, Knaus WA, Lynn J: Study population in SUPPORT: Patients, surrogates, and physicians. J Clin Epidemiol 434:11S-28S, 1990 (suppl) 38. Phillips RS, Knaus WA: Patient characteristics in SUPPORT: Sociodemographics, admission diagnosis, comorbidities, and acute physiology score. J Clin Epidemiol 434:29S-31S, 1999 (suppl) 39. Kreling B, Wu AW, Lynn J: Survey methods for seriously ill hospitalized adults: Practical lessons from SUPPORT. J Am Geriatr Soc 48:S168-S175, 2000 (suppl)[Medline] 40. Schaeffer JL: Analysis of incomplete multivariate data. London, UK, Chapman and Hall, 1997 41. Youngner SJ, Murphy DJ, Lynn J. Decision making in SUPPORT: Sentinel decisions. J Clin Epidemiol 43:67S-71S, 1990 (suppl) 42. Keene RA, Cullen DJ: Therapeutic Intervention Scoring System: Update 1983. Crit Care Med 11:1-3, 1983[Medline]
43. Knaus WA, Harrell FE, Lynn J, et al: The SUPPORT prognostic model: Objective estimates of survival for seriously ill hospitalized adults. Ann Intern Med 122:191-203, 1995 44. Cox DR: Regression models and life-tables with discussion. J R Stat Soc 34:187-220, 1972 45. Therneau T, Humiton SA: rhDNASE as an example of recurrent event analysis. Stat Med 16:2029-2047, 1997[CrossRef][Medline] 46. Anderson PK, Gill RD: Cox regressions model for counting processes: A large sample study. Ann Stat 10:1100-1120, 1982 47. Degner LF, Sloan JA: Decision making during serious illness: What role do patients really want to play? J Clin Epidemiol 45:941-950, 1992[CrossRef][Medline] 48. Dawson NV, Rose JH, Thomas C, et al: Survival estimates and accuracy of prognostic predictions for cancer patients by physician specialty. J Gen Intern Med 13:18, 1998 (suppl; abstr) 49. Rose JH, Haug MR: Communication and the cancer patient: Information and truth [book review]. J Ethics Law Aging 5:85-87, 1999 50. Miyaji NT: The power of compassion: Truth-telling among American doctors in the care of dying patients. Soc Sci Med 36:249-264, 1993 51. Miller SM: Monitoring versus blunting styles of coping with cancer influence the information patients want and need about their disease. Cancer 76:167-177, 1995[CrossRef][Medline] 52. Ong LML, Visser MRM, Van Zuuren FJ, et al: Cancer patients' coping styles and doctor-patient communication. Psycho-Oncol 8:155-166, 1999[CrossRef][Medline] 53. Sachs GA, Ahronheim JC, Rhymes JA, et al: Good care of the dying. J Am Geriatr Soc 43:553-562, 1995[Medline] 54. Chen H, Haley WE, Robinson BE, et al: Decisions for hospice care in patients with advanced cancer. J Am Geriatr Soc 51:789-797, 2003[CrossRef][Medline]
55. Randal J: Serving those for whom cure is no longer feasible (news). J Natl Cancer Inst 88:1018-1020, 1996 56. Lynn J, Arkes HR, Stevens M: Rethinking fundamental assumptions: SUPPORT's implications for future reform. J Am Geriatr Soc 48:S214-S221, 2000[Medline] Submitted June 13, 2003; accepted June 19, 2004. Related Article
Related Editorial
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||
|
TOP
ABSTRACT
INTRODUCTION
