|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 2 (January 15), 2004: pp. 315-321 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.08.136 Trends in the Aggressiveness of Cancer Care Near the End of LifeFrom the Divisions of Population Sciences and Psychosocial Oncology and Palliative Care, Department of Medical Oncology, Dana-Farber Cancer Institute; Department of Health Care Policy, Harvard Medical School; and Division of General Medicine and Department of Psychiatry, Brigham & Women’s Hospital, Boston, MA Address reprint requests to Craig C. Earle, MD, MSc, Center for Outcomes and Policy Research, Dana-Farber Cancer Center, 44 Binney St, 454-STE 21-24, Boston, MA 02115; e-mail: craig_earle{at}dfci.harvard.edu.
PURPOSE: To characterize the aggressiveness of end-of-life cancer treatment for older adults on Medicare, and its relationship to the availability of healthcare resources. PATIENTS AND METHODS: We analyzed Medicare claims of 28,777 patients 65 years and older who died within 1 year of a diagnosis of lung, breast, colorectal, or other gastrointestinal cancer between 1993 and 1996 while living in one of 11 US regions monitored by the Surveillance, Epidemiology, and End Results Program. RESULTS: Rates of treatment with chemotherapy increased from 27.9% in 1993 to 29.5% in 1996 (P = .02). Among those who received chemotherapy, 15.7% were still receiving treatment within 2 weeks of death, increasing from 13.8% in 1993 to 18.5% in 1996 (P < .001). From 1993 to 1996, increasing proportions of patients had more than one emergency department visit (7.2% v 9.2%; P < .001), hospitalization (7.8% v 9.1%; P = .008), or were admitted to an intensive care unit (7.1% v 9.4%; P = .009) in the last month of life. Although fewer patients died in acute-care hospitals (32.9% v 29.5%; P < .001) and more used hospice services (28.3% v 38.8%; P < .001), an increasing proportion of patients who received hospice care initiated this service only within the last 3 days of life (14.3% v 17.0%; P = .004). Black patients were more likely than white patients to experience aggressive intervention in nonteaching hospitals but not in teaching hospitals. Greater local availability of hospices was associated with less aggressive treatment near death on multivariate analysis. CONCLUSION: The treatment of cancer patients near death is becoming increasingly aggressive over time.
Issues pertaining to end-of-life cancer care affect a large number of Americans each year [1]. The last decade has seen an expansion in the therapeutic possibilities for patients with advanced cancer [2-6]. At the same time, there has been a recognition of the importance of high-quality end-of-life care, including end-of-life decision making, pain and symptom management, psychosocial support, and hospice care [7]. Through literature review [8-12] and focus groups with patients, family members, and physicians, we have previously identified several indicators of aggressive care near the end of life: intensive use of chemotherapy, low rates of hospice use, and interventions resulting in emergency room (ER) visits, hospitalization, or intensive care unit (ICU) admissions [13]. The objectives of this study were to observe these markers of aggressive care in a cohort of patients with advanced solid tumors, to evaluate the trends in the aggressiveness of care over time in a population-based cohort, and to assess their association with the availability of related healthcare resources such as hospice services.
Two data sources were used: data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) registries, and Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration) Medicare claims data. The eleven tumor registries participating in the SEER program capture approximately 97% of the incident cases [15], covering a representative sample [16] of approximately 14% of the American population [17]. Registries collect data on each patient’s age, sex, race and ethnicity, cancer site, stage, histology, date of diagnosis, and the date and cause of death. Medicare claims for inpatient and outpatient care, physician and laboratory billings, as well as bills for home health and hospice care have been linked to SEER for patients age 65 years and older [18]. Census-level sociodemographic data have also been linked to these patients.
Cohort Selection To evaluate trends in care over time, it was necessary to restrict the cohort to patients who had a survival time of less than 1 year so that patients dying in later years were not ones who had survived longer than those dying in the earlier years. For sensitivity analyses, all analyses were repeated on cohorts restricted to those who presented with metastatic disease and patients younger than age 75 years, and were stratified by cancer subtype. The results of each of these analyses were similar and are not reported.
Definitions of Explanatory Variables Available resources in the local community were assessed using the Health Care Service Area (HCSA; groupings of metropolitan statistical areas defined by observed patient flow patterns in Medicare [25]) as the unit of analysis. Data from the National Center for Health Workforce Information and Analysis’ Area Resource File were used to calculate the following HCSA characteristics: the per capita density of physicians, medical specialists, and radiation oncologists derived from the American Medical Association Master File, and the density of hospitals, teaching hospitals, hospital beds, hospitals with oncology services, and hospices from the County Hospital File. The 77 HCSAs comprising the SEER areas were then grouped into tertiles for analysis.
Statistical Analysis
Univariate analyses comparing characteristics of patients and the care received were carried out using t tests for continuous variables and
Rates of chemotherapy use increased during the 4 years from 27.9% in 1993 to 29.5% in 1996 (P = .02). Table 1 shows the characteristics of the 28,777 patients who met eligibility criteria, broken down by year of death.
Measures of Care Near the End of Life Among 8,155 patients who received chemotherapy, the average number of days between the start of the last unique chemotherapy regimen and death decreased from 140.5 to 127.7 (P = .008; Table 2), whereas the interval between the last dose of chemotherapy and death steadily decreased from 71 days in 1993 to 65.3 days in 1996 (P < .001). The mean duration of the last course of chemotherapy was 61 days and did not change over time. The mean time between the last dose of chemotherapy and entry into hospice was 58.1 days, but the distribution was skewed, with a median of 37 days. It decreased steadily from 61.7 to 55.1 days over the 4-year study period (P < .001).
There were more ER visits in the last month of life in 1996 than there were in 1993, increasing from 0.40 to 0.46 visits per person (P < .001); however, there were no discernible trends in the numbers of hospital admissions or inpatient days. Despite this, the number of ICU admissions between 1993 and 1996 increased from 0.10 to 0.12 per person (P < .001), as did the average number of days in the ICU (0.52 to 0.61; P < .001). Some opposing trends also were observed, with more patients using hospice services over time (28.3% in 1993 v 38.8% in 1996; P < .001), and fewer patients dying in acute-care hospitals (32.9% v 29.5%; P < .001). Patients who were treated with chemotherapy were more likely than those not receiving chemotherapy to die in hospital (34.9% v 29.0%; P < .001) however, and, if they were admitted to hospice, those who had been treated with chemotherapy were admitted closer to death (26.6 v 35.2 days; P < .001). These figures did not change significantly over time.
Indicators of Aggressive Care
These indicators of aggressive care were all significantly correlated with each other using Spearman rank testing. Thirty percent of patients had at least one indicator of aggressive care, and more than half of those had more than one indicator of aggressive care, increasing steadily from 29.0% to 32.3% during the 4-year study period (P < .001). Patients with an indicator of intense chemotherapy use were significantly more likely also to have an ER visit, hospitalization, or ICU admission (P = .001). Hospice use was inversely correlated with indicators of aggressive care. Forty-two percent of patients without any indicator of aggressive care were admitted to hospice. In contrast, only 17% of patients experiencing any of these indicators used hospice services (P < .001): 22% of patients with one indicator, 12% with two indicators, 11% with three indicators, and less than 10% with more than three indicators entered hospice (Fig 1). Furthermore, those patients with an indicator of aggressive care who entered hospice were significantly more likely to do so within 3 days of death (35.6% v 12.1%; P < .001).
Multivariate Analyses Logistic regression analysis found year of death to be a significant independent predictor of the likelihood of having one or more indicators of aggressive care after adjustment for other explanatory variables (odds ratio [OR], 1.06 for each increasing year; 95% CI, 1.02 to 1.10; Table 4). Increased age decreased the odds of experiencing an indicator of aggressive care (OR, 0.98 for each year; 95% CI, 0.97 to 0.99), as did female sex (OR, 0.80; 95% CI, 0.73 to 0.87). Comorbidity was a predictor of aggressive care (OR, 1.14; 95% CI, 1.06 to 1.23). On additional examination, patients with comorbid conditions who were treated with chemotherapy were more likely than those without comorbidity to have ER visits and admissions to hospital in the last month of life, even when these measures were restricted to cancer-related occurrences. Both care in a teaching hospital (OR, 1.24; 95% CI, 1.12 to 1.38) and the regional characteristic of having more teaching hospitals per capita (OR, 1.10; 95% CI, 1.04 to 1.17) independently predicted more aggressive care. Black race increased the likelihood of receiving more aggressive care; however, there was an interaction with whether the patient received care in a teaching hospital such that the effect of race was seen specifically in patients who only received care in a nonteaching setting (OR, 1.25; 95% CI, 1.01 to 1.55). A higher density of hospices in an HCSA was associated with a significant, independent decrease in the likelihood of experiencing an indicator of aggressive care (OR, 0.93; 95% CI, 0.88 to 0.98). There were no discernible associations on univariate or multivariate evaluation between socioeconomic status, cancer type, stage at presentation, or geographic region and the aggressiveness of care. There was no clear association of the per capita number of physicians, medical specialists, or hospitals available with the aggressiveness of care, and HMO penetration as a regional characteristic did not affect the aggressiveness of care in this population of exclusively fee-for-service patients.
In this study of terminally ill patients with cancer, we found evidence of an increasing propensity to use aggressive interventions near death during a 4-year period in the mid-1990s, a trend associated with increased ER visits, hospitalizations, and ICU admissions. Although the absolute changes observed from year to year were small, the relative changes were remarkable and the trends were consistent. Interestingly, regions with greater availability of hospice services had less aggressive care near the end of life, suggesting that efforts to enhance hospice access may temper the use of more aggressive interventions for patients in this advanced phase of their illness. The effect of regional resources on practice patterns is particularly intriguing because it suggests immediate policy implications. Rates of hospice use have been found to be more closely related to the availability of hospice services than to patient preference [27,28]. Others have found that hospice use is higher in areas with fewer hospital beds per capita, more physicians, and areas with higher HMO enrollment [29]. The likelihood of dying with cancer in a hospital has also previously been reported to be increased in men, blacks, and the poor [27,30,31]. The relationship between comorbidity and these measures of aggressiveness of care is complex. Patients with comorbid conditions were less likely to receive chemotherapy, but if they did, their comorbidity did not seem to mediate the aggressiveness with which chemotherapy was given, and perhaps as a result, they were more likely to end up in the ER or be admitted to hospital in the last month of life. Concurrent with an increase in some markers of aggressive care, there was also an increase in hospice use and decrease in hospital deaths. However, these results do not necessarily conflict. Increasingly aggressive care appears to be superimposed on a secular trend of increasing hospice use whether or not patients receive aggressive care. Furthermore, although more patients are being admitted to hospice, more of these patients are being admitted close to death. This suggests that hospice is being used in these cases to manage death itself, rather than palliating the disease. Since the end of this study there have been several relevant statements from professional organizations emphasizing the importance of end-of-life and palliative care [32,33]. As a result, continued monitoring of these trends is needed. Several authors have outlined concepts related to the quality of end-of-life care, and many, though not all [34], have included aggressiveness of care as a domain [35,36], at least to the extent of discussing resuscitation and life support [37]. However, treatment decisions for terminally ill older adults with advanced solid tumors are complex, making quality-of-care assessment difficult in this setting. Choices about aggressive interventions involve not only access to hospice, but also patient and family attitudes about hospice, chemotherapy, and the relationship with their primary and specialist physicians. Unfortunately, we are not able to observe these preferences in administrative data. Our study has some limitations. To create an unbiased cohort in a steady-state over the study period, we had to restrict eligibility to only those patients with a relatively short survival time. We could not determine whether the events we observed reflect physician attitudes, availability and access to resources, or patient preferences. It is not possible for the clinician in real time to know exactly when a patient will die [38,39], although we would not expect prognostic abilities to be systematically worsening over time. These results in elderly patients with advanced solid tumors in a fee-for-service environment may not translate to other populations or settings [40], and the analysis relies on the accuracy of billing data, which were not collected for health services research. Strengths include the use of population-based data from multiple regions for a relatively large cohort. Patients all had Medicare insurance providing consistent financial incentives for patients and providers, although we did not have information on patients’ supplemental insurance. This study is the first to indicate that the treatment of advanced cancer patients is becoming increasingly aggressive and that aggressive treatment is continuing later in life. Whether such care has positive effects on survival or quality of life is questionable and deserves additional investigation. There is a growing need for policy makers to reframe the Medicare hospice benefit to make combined cancer treatment and hospice care financially feasible to allow patients to have access to potentially beneficial treatments for their cancer without sacrificing excellent end-of-life care [41].
The authors indicated no potential conflicts of interest.
Supported by a grant from the National Cancer Institute (CA 91753-02). Presented at the 39th Annual Meeting of the American Society of Clinical Oncology, May 31-June 3, 2003, Chicago, IL. Authors’ disclosures of potential conflicts of interest are found at the end of this article.
1. Hoyert DL, Kochanek KD, Murphy SL: Deaths: Final data for 1997. Natl Vital Stat Rep 47: 1-104, 1999[Medline]
2. Cormier Y, Eisenhauer E, Muldal A, et al: Gemcitabine is an active new agent in previously untreated extensive small cell lung cancer (SCLC): A study of the National Cancer Institute of Canada Clinical Trials Group. Ann Oncol 5: 283-285, 1994 3. Comis RL, Friedland DM: New chemotherapy agents in the treatment of advanced non-small cell lung cancer: An update including data from the Seventh World conference on Lung Cancer. Lung Cancer 12: S63-S99, 1995 (suppl 2) 4. Steward WP, Dunlop DJ: New drugs in the treatment of non-small cell lung cancer. Ann Oncol 6: S49-S54, 1995 5. Punt CJ: New drugs in the treatment of colorectal cancer. Cancer 83: 679-689, 1998[CrossRef][Medline] 6. Bleiberg H: Continuing the fight against advanced colorectal cancer: New and future treatment options. Anticancer Drugs 9: 18-28, 1998[CrossRef][Medline] 7. Institute of Medicine: Improving Palliative Care for Cancer. Washington, DC, National Academy Press, 2001, p 1 8. Boyce N, McNeil J, Graves D, et al: Quality and outcome indicators for acute health care services. Canberra, Australia, Australian Government Publishing Service. http://www.health.gov.au/pubs/qualrprt/qohmepge.htm 9. Kurowski B: Cancer carve outs, specialty networks, and disease management: A review of their evolution, effectiveness, and prognosis. Am J Manag Care 4: SP71-SP89, 1998 10. Hillner BE, Smith TJ: The quality of cancer care: Does the literature support the rhetoric? Report to the National Cancer Policy Board. http://www.gyndoctor.com/quality%20cancer%20care1.pdf
11. Mandelblatt JS, Ganz PA, Kahn KL: Proposed agenda for the measurement of quality-of-care outcomes in oncology practice. J Clin Oncol 17: 2614-2622, 1999 12. Smith TJ, Desch CE, Hillner BE. The quality of cancer care: Models of excellence—Report to the National Cancer Policy Board. http://www.iom.edu/includes/dbfile.asp?id=12801 13. Mortenson LE: How to judge the cancer services benefit component of your health insurance plan. Cancer 82: 2061-2067, 1998[CrossRef][Medline]
14. Earle CC, Park ER, Lai B, et al: Identifying potential indicators of the quality of end-of-life cancer care from administrative data. J Clin Oncol 21: 1133-1138, 2003 15. Zippin C, Lum D, Hankey BF: Completeness of hospital cancer case reporting from the SEER program of the National Cancer Institute. Cancer 76: 2343-2350, 1995[CrossRef][Medline] 16. Nattinger AB, McAuliffe TL, Schapira MM: Generalizability of the Surveillance, Epidemiology, and End Results Registry population: Factors relevant to epidemiologic and health care research. J Clin Epidemiol 50: 939-945, 1997[CrossRef][Medline] 17. Ries LAG, Kosary CL, Hankey BF, et al: SEER Cancer Statistics Review, 1973-1994. Bethesda, MD, National Cancer Institute, NIH publication 97-2789, 1997 18. Potosky AL, Riley GF, Lubitz JD, et al: Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care 31: 732-748, 1993[Medline] 19. American Cancer Society: Cancer statistics 2000. CA Cancer J Clin 50: 7-33, 2000[Abstract]
20. Krieger N: Overcoming the absence of socioeconomic data in medical records: Validation and application of a census-based methodology. Am J Public Health 82: 703-710, 1992 21. Deyo RA, Cherkin DC, Ciol MA: Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. Clin Epidemiol 45: 613-619, 1992[CrossRef][Medline] 22. Charlson ME, Pompei P, Ales KL, et al: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 40: 373-383, 1987[CrossRef][Medline] 23. Klabunde CN: Development of a comorbidity index using physician claims data. J Clin Epidemiol 53: 1258-1267, 2000[CrossRef][Medline]
24. Earle CC, Tsai JS, Gelber RD, et al: Effectiveness of palliative chemotherapy for advanced lung cancer: Instrumental variable and propensity analysis. J Clin Oncol 19: 1064-1070, 2001 25. Makuc DM, Haglund B, Ingram DD, et al: Vital and Health Statistics: Health Service Areas for the United States—Series 2: Data Evaluation and Methods Research 112. Washington, DC, Department of Health and Human Services, DHHS publication (PHS) 92-1386, 1991
26. Kiefe CI, Weissman NW, Allison JJ, et al: Identifying achievable benchmarks of care: Concepts and methodology. Int J Qual Health Care 10: 443-447, 1998
27. Bruera E, Russell N, Sweeney C, et al: Place of death and its predictors for local patients registered at a comprehensive cancer center. J Clin Oncol 20: 2127-2133, 2002 28. Virnig BA, Fisher ES, McBean MA, et al: Hospice use in Medicare managed care and fee-for-service systems. Am J Manag Care 7: 777-786, 2001[Medline] 29. Virnig BA, McBean M, Kind S, et al: Hospice use before death. Med Care 40: 73-78, 2002[CrossRef][Medline] 30. Virnig BA, Kind S, McBean M, et al: Geographic variation in hospice use prior to death. J Am Geriatr Soc 48: 1117-1125, 2000[Medline] 31. Gallo WT, Baker MJ, Bradly EH: Factors associated with home versus institutional death among cancer patients in Connecticut. J Am Geriatr Soc 49: 771-777, 2001[CrossRef][Medline] 32. Institute of Medicine: Approaching Death. Washington, DC, National Academy Press, 1997 33. American Society of Clinical Oncology: Cancer care during the last phase of life. J Clin Oncol 16: 1986-1996, 1998[Abstract]
34. Patrick DL, Curtis JR, Engelberg RA, et al: Measuring and improving the quality of dying and death. Ann Intern Med 139: 410-415, 2003 35. Lynn J: Measuring quality of care at the end of life: A statement of principles. J Am Geriatr Soc 45: 528-530, 1997[Medline]
36. Singer PA, Martin DK, Keiner M: Quality end-of-life care: Patient’s perspectives. JAMA 281: 163-168, 1999 37. Emanuel EJ, Emanuel LL: The promise of a good death. Lancet 351: 21-29, 1998 (suppl II) 38. Senn S, Harrell F: On wisdom after the event. J Clin Epidemiol 50: 749-751, 1997[CrossRef][Medline] 39. Teno JM, Coppola KM: For every numerator, you need a denominator: A simple statement but key to measuring the quality of care of the dying. J Pain Symptom Manage 17: 109-113, 1999[CrossRef][Medline]
40. McCarthy EP, Burns RB, Ngo-Metzger Q, et al: Hospice use among Medicare managed care and fee-for-service patients dying with cancer. JAMA 289: 2238-2245, 2003 41. Schapiro R, Byock I, Parker S, et al: Living and dying well with cancer: Successfully integrating palliative care and cancer treatment. Robert Wood Johnson Foundation, 2003 Submitted August 20, 2003; accepted November 7, 2003. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 analyses for categorical variables. The Cochran-Armitage test was used to evaluate time trends for categorical variables, whereas Spearman rank-order correlations were sought for continuous variables. On the basis of variables found to be significantly associated with the composite measure of aggressiveness of care on univariate results, logistic regression was used to examine the simultaneous influence of multiple predictors. Significant explanatory variables were identified through forward selection, and interaction terms between the significant variables were investigated further. No variable was forced in. All statistical analyses were performed with Statistical Analysis software, version 8.1 for Windows (SAS/STAT User’s Guide; SAS Institute, Cary, NC, 1990). 
